The document provides 19 mistakes to avoid when configuring HA and DRS in a vSphere environment. Some key mistakes include: not planning for hardware evolution which can prevent vMotion between newer and older hosts; neglecting host isolation response settings which determines what happens to VMs on an isolated host; overdoing reservations, limits, and affinities which can constrain HA and DRS calculations; and being too liberal with the DRS migration threshold which can result in unnecessary migrations. The document emphasizes understanding how DRS evaluates resource utilization and models migrations to balance cluster load.
VMworld US 2011 - Avoiding the 16 Biggest HA & DRS Configuration MistakesConcentrated Technology
Everyone thinks HA and DRS are wonderful technologies. Yet both can be notoriously dangerous when misconfigured. Make mistakes with either, and they will take down your virtual infrastructure more quickly than you can imagine. VMware vExpert Greg Shields has collected 16 of the biggest mistakes he has seen in his consulting travels. In this not-to-be-missed session, Greg shares all, along with simple solutions that will keep your virtual environment healthy. Be sure not to miss this session. You will want to remote your systems immediately afterward to see if you have made these mistakes too.
The document discusses monitoring ESX server performance and provides tips for troubleshooting performance issues. It outlines a 10 step process for analyzing performance that includes checking VMware tools, host and guest CPU usage, memory usage and swapping, network throughput and packet drops, IOPS, storage latency, and VM CPU utilization. The presentation aims to provide a structured approach for identifying and addressing the root causes of performance problems.
How DreamHost builds a Public Cloud with OpenStackCarl Perry
This document summarizes DreamHost's presentation on how they built a public cloud using OpenStack. Some key points:
- DreamHost is using OpenStack for compute, storage, and networking in their public cloud offering called DreamCompute.
- For storage, they chose Ceph which provides shared, scalable block and object storage.
- Their network architecture uses 10Gb switches in a spine-leaf topology with logical networking software for tenant isolation.
- Automation is key to managing the cloud infrastructure and providing services to customers.
- DreamHost discussed the considerations and challenges in building the cloud such as scalability, speed, monitoring, security and cost effectiveness.
Virtualization Manager 5.0 – Now with Hyper-V Support!SolarWinds
For more information on Virtualization Manager, visit: http://www.solarwinds.com/virtualization-manager.aspx
Watch this webcast: http://www.solarwinds.com/resources/webcasts/virtualization-manager-50-now-with-hyperv-support.html
Whether you have a Hyper-V virtual environment, VMware, or both – Virtualization Manager now has you covered. Watch SolarWinds virtualization experts Brian Radovich and Robbie Wright as we discuss the key areas for managing a Hyper-V virtual environment.
• How to manage performance on a shared virtual infrastructure
• Building out a proactive capacity plan
• Tracking and reporting on virtual configurations and drift
• Living in a multi-hypervisor world!
Also during this webcast we demonstrate key technologies from SolarWinds that help to conquer these challenges and ensure success in virtual environments.
The introduction of DateTieredCompactionStrategy in late 2014 was a significant step forward in providing a viable compaction strategy for time series data, especially time series data that will be TTL'd out. DateTieredCompactionStrategy's introduction was met with genuine excitement, and its rapid adoption is testament to developers' and operators' desire to have data compacted in a way that better matches their write patterns.
However, DateTieredCompactionStrategy's features come with significant limitations. This talk will review our real world benchmarking and use cases for DTCS as a vehicle to discuss the implications of DateTieredCompactionStrategy on operational tasks such as repair, read-repair, bootstrapping, and especially DR recovery scenarios, and it will also discuss how those various limitations lead us to proposing an operations-friendly alternative to DateTieredCompactionStrategy.
Squeeze Maximum Performance from your Hosting PlatformSiteGround.com
The presentation covers useful insights and benchmark tests on how the performance (and security) of PHP/MySQL based applications can be significantly improved through different tweaks done at server administration level (Linux/Apache). It also examines several different types of hosting platforms: dedicated, virtual/cloud and shared, and how they can influence the CMS application speed and security.
A presentation delivered by SiteGround CEO at CMS Expo - Chicago, May 8-10 2012.
STO7535 Virtual SAN Proof of Concept - VMworld 2016Cormac Hogan
This document provides an overview of tools that can help administrators successfully conduct a Virtual SAN proof of concept. It discusses the Virtual SAN Health Check plugin, capacity views, performance service, HCIbench, and Virtual SAN Observer for monitoring and validating Virtual SAN configurations. Validation scenarios covered include successfully deploying Virtual SAN, deploying VMs on VSAN storage, VM availability during host and storage failures, and measuring rebuild activity.
VMworld US 2011 - Avoiding the 16 Biggest HA & DRS Configuration MistakesConcentrated Technology
Everyone thinks HA and DRS are wonderful technologies. Yet both can be notoriously dangerous when misconfigured. Make mistakes with either, and they will take down your virtual infrastructure more quickly than you can imagine. VMware vExpert Greg Shields has collected 16 of the biggest mistakes he has seen in his consulting travels. In this not-to-be-missed session, Greg shares all, along with simple solutions that will keep your virtual environment healthy. Be sure not to miss this session. You will want to remote your systems immediately afterward to see if you have made these mistakes too.
The document discusses monitoring ESX server performance and provides tips for troubleshooting performance issues. It outlines a 10 step process for analyzing performance that includes checking VMware tools, host and guest CPU usage, memory usage and swapping, network throughput and packet drops, IOPS, storage latency, and VM CPU utilization. The presentation aims to provide a structured approach for identifying and addressing the root causes of performance problems.
How DreamHost builds a Public Cloud with OpenStackCarl Perry
This document summarizes DreamHost's presentation on how they built a public cloud using OpenStack. Some key points:
- DreamHost is using OpenStack for compute, storage, and networking in their public cloud offering called DreamCompute.
- For storage, they chose Ceph which provides shared, scalable block and object storage.
- Their network architecture uses 10Gb switches in a spine-leaf topology with logical networking software for tenant isolation.
- Automation is key to managing the cloud infrastructure and providing services to customers.
- DreamHost discussed the considerations and challenges in building the cloud such as scalability, speed, monitoring, security and cost effectiveness.
Virtualization Manager 5.0 – Now with Hyper-V Support!SolarWinds
For more information on Virtualization Manager, visit: http://www.solarwinds.com/virtualization-manager.aspx
Watch this webcast: http://www.solarwinds.com/resources/webcasts/virtualization-manager-50-now-with-hyperv-support.html
Whether you have a Hyper-V virtual environment, VMware, or both – Virtualization Manager now has you covered. Watch SolarWinds virtualization experts Brian Radovich and Robbie Wright as we discuss the key areas for managing a Hyper-V virtual environment.
• How to manage performance on a shared virtual infrastructure
• Building out a proactive capacity plan
• Tracking and reporting on virtual configurations and drift
• Living in a multi-hypervisor world!
Also during this webcast we demonstrate key technologies from SolarWinds that help to conquer these challenges and ensure success in virtual environments.
The introduction of DateTieredCompactionStrategy in late 2014 was a significant step forward in providing a viable compaction strategy for time series data, especially time series data that will be TTL'd out. DateTieredCompactionStrategy's introduction was met with genuine excitement, and its rapid adoption is testament to developers' and operators' desire to have data compacted in a way that better matches their write patterns.
However, DateTieredCompactionStrategy's features come with significant limitations. This talk will review our real world benchmarking and use cases for DTCS as a vehicle to discuss the implications of DateTieredCompactionStrategy on operational tasks such as repair, read-repair, bootstrapping, and especially DR recovery scenarios, and it will also discuss how those various limitations lead us to proposing an operations-friendly alternative to DateTieredCompactionStrategy.
Squeeze Maximum Performance from your Hosting PlatformSiteGround.com
The presentation covers useful insights and benchmark tests on how the performance (and security) of PHP/MySQL based applications can be significantly improved through different tweaks done at server administration level (Linux/Apache). It also examines several different types of hosting platforms: dedicated, virtual/cloud and shared, and how they can influence the CMS application speed and security.
A presentation delivered by SiteGround CEO at CMS Expo - Chicago, May 8-10 2012.
STO7535 Virtual SAN Proof of Concept - VMworld 2016Cormac Hogan
This document provides an overview of tools that can help administrators successfully conduct a Virtual SAN proof of concept. It discusses the Virtual SAN Health Check plugin, capacity views, performance service, HCIbench, and Virtual SAN Observer for monitoring and validating Virtual SAN configurations. Validation scenarios covered include successfully deploying Virtual SAN, deploying VMs on VSAN storage, VM availability during host and storage failures, and measuring rebuild activity.
Cassandra Summit 2015: Real World DTCS For OperatorsJeff Jirsa
Real World DTCS For Operators
The introduction of DateTieredCompactionStrategy in late 2014 was a significant step forward in providing a viable compaction strategy for time series data, especially time series data that will be TTL'd out. DateTieredCompactionStrategy's introduction was met with genuine excitement, and its rapid adoption is testament to developers' and operators' desire to have data compacted in a way that better matches their write patterns.
However, DateTieredCompactionStrategy's features come with significant limitations. This talk will review our real world benchmarking and use cases for DTCS as a vehicle to discuss the implications of DateTieredCompactionStrategy on operational tasks such as repair, read-repair, bootstrapping, and especially DR recovery scenarios, and it will also discuss how those various limitations lead us to proposing an operations-friendly alternative to DateTieredCompactionStrategy.
A day in the life of a VSAN I/O - STO7875Duncan Epping
This document provides an overview and summary of a VMworld session about Virtual SAN I/O. The session covers Virtual SAN concepts, the I/O flow of reads and writes in Virtual SAN, failure scenarios and how Virtual SAN handles them, and new features like deduplication and compression. The document includes diagrams demonstrating how data is distributed and replicated across hosts in a Virtual SAN cluster. It also provides details on how reads, writes, and failures are handled at a technical level in Virtual SAN. In the conclusion, it recommends three ways for attendees to get started with Virtual SAN: a hands-on lab, 60-day free evaluation, or working with a VMware partner on an assessment.
Virtualizing Tier One Applications - VarrowAndrew Miller
This document provides best practices for virtualizing mission critical applications like Exchange and SQL Server. It discusses the top 10 myths about virtualizing business critical applications and provides the truths. It then discusses best practices for virtualizing Exchange, including starting simple, licensing, storage configuration, and high availability options. For SQL Server, it covers starting simple, licensing, storage configuration, migrating, and database best practices. It also discusses tools that can be used for database performance analysis when virtualized like Confio IgniteVM and vCenter Operations.
- The document provides best practices for deploying and managing VMware vSphere environments. It covers topics such as ESXi host deployment, virtual machine deployment, vSphere HA clusters, vSphere networking, snapshots, security, vCenter deployment, backups, technical support, and housekeeping.
- Key recommendations include using compatible hardware, stable versions, host profiles, latest virtual hardware, syspreped templates, DRS automation, separate infrastructure and VM traffic, vSphere distributed switches, meaningful snapshots, external databases for vCenter, backups, and change management.
The have no fear guide to virtualizing databasesSolarWinds
When it comes to a successful database virtualization journey, there are things you must know before you start. In this presentation you will:
-Review terms and concepts for VMware, by far the most common virtualization platform
-Examine how to use vSphere (the VMware admin console)
-Explore the differences between virtual and physical host metrics a
-Learn to overcome the shortcomings of virtualizing your database environment
vSphere APIs for performance monitoringAlan Renouf
The document discusses vSphere APIs for performance monitoring and provides an overview of useful performance metrics such as CPU, memory, disk, and network. It motivates the need for these APIs by demonstrating how charts and visualizations can help identify performance issues. The agenda covers common useful statistics, how vSphere retrieves performance data, and how to access the data programmatically using tools like PowerCLI and Java/C#.
Abstract:
Reactive applications need to be able to respond to demand, be elastic and ready to scale up, down, in and out—taking full advantage of mobile, multi-core and cloud computing architectures—in real time.
In this talk we will discuss the guiding principles making this possible through the use of share-nothing and non-blocking designs, applied all the way down the stack. We will learn how to deliver systems that provide reactive supply to changing demand.
I gave this talk at React Conf 2014 in London. Recording available here: https://www.youtube.com/watch?v=mBFdj7w4aFA
Presentation architecting a cloud infrastructurexKinAnx
This document provides an agenda and overview for a session on architecting a cloud infrastructure. The agenda includes introductions, gathering requirements, sizing and scaling, host design, vCenter design, cluster design, networking and storage considerations. It emphasizes the importance of gathering requirements from customers and conceptualizing the design based on those requirements. It also discusses various design considerations and best practices for each component of a cloud infrastructure.
SaltStack can be used to automate and orchestrate the provisioning of virtual machines on VMware ESXi 6.0. It implements the VMware APIs to allow defining VM profiles and templates that specify VM configurations, and then uses Salt commands to rapidly deploy new VMs from templates with customized configurations. Open-VM tools must be installed on templates to enable customizing VMs, such as setting the network configuration. Salt files define VM profiles and provider credentials, separating configuration from deployment logic for flexibility and reusability.
How we collaborated with the CentOS and Xen projects to build a next-generation platform at Go Daddy. Discussion of the design considerations, infrastructure, succes stories and challenges of this paradigm change
This document is a slide deck presentation about high availability (HA) and distributed resource scheduler (DRS) gotchas that can negatively impact virtual infrastructure. It discusses common issues administrators encounter with vMotion capabilities as hardware is refreshed over time. It also explains how DRS and HA work together to optimize resource utilization and provide fault tolerance through automated workload balancing and live migration of virtual machines during host failures. The presentation emphasizes properly configuring clusters, resource pools, and rules to allow DRS and HA to fully leverage vMotion's capabilities.
VMworld 2011 Review: Preparing for vSphere 5 with Virtualization ManagerSolarWinds
For more information on Virtualization Manager, visit: http://www.solarwinds.com/virtualization-manager.aspx
Watch this webcast: http://www.solarwinds.com/resources/webcasts/vmworld-us-takeaways-insights-and-best-practices.html
During this Geek Speak webcast, you will learn why Virtualization Manager won the Best of VMworld 2011 Gold Award for Virtualization Management and how to solve the issues affecting you the most.
Slides from the second meeting of the Toronto High Scalability Meetup @ http://www.meetup.com/toronto-high-scalability/
-Basics of High Scalability and High Availability
-Using a CDN to Achieve 99% Offload
-Caching at the Code Layer
Virtualization allows computer resources to be abstracted and shared. It involves simulating hardware to run virtual machines that behave like physical computers. A hypervisor manages virtual machines and allocates resources to them. Virtualization provides benefits like efficient resource use, increased availability, disaster recovery, and on-demand scaling. While it adds complexity, virtualization is commonly used to virtualize desktops, run specific programs, create test environments, design private clouds, and utilize public clouds. Major vendors provide virtualization software and cloud services.
24 Hours of PASS, Summit Preview Session: Virtual SQL Server CPUsDavid Klee
This document discusses virtual CPUs and CPU architecture. It begins by explaining how hypervisor resource queues work and how requests for CPU, memory, storage and networking are placed in queues. It then covers physical CPU architecture including cores, sockets, NUMA and memory locality. It discusses how virtual CPUs are scheduled by the hypervisor and ways to measure scheduling pressure. Finally, it provides recommendations for right-sizing virtual machines and balancing workloads to reduce scheduling delays.
This document provides an overview of implementing affordable disaster recovery with Hyper-V and multi-site clustering. It discusses what constitutes a disaster, the key components needed which are a storage mechanism, replication mechanism, and target servers/cluster. It also covers clustering history, what a cluster is, and the important concept of quorum which determines a cluster's existence through voting of its members.
The popularity of Virtual SAN is growing daily. Server admins are finally free to aggregate storage in their servers to create a shared storage system that scales with their compute needs. The underlying key to making it all work is networking. All Virtual SAN data flows through it, and correct selection and configuration of networking components will mean the difference between disruptive success or dramatic failure. This session will give deep insight in the do's and don'ts of Virtual SAN networking. Best practices for physical and virtual switch configuration and performance testing will be discussed. Virtual SAN 5.5 and 6.0 will be covered, and the networking differences discussed. Methods of troubleshooting network issues will be covered. For those configuring a Virtual SAN network for the first time, for labs or enterprise scale, this session is a must-see.
WinConnections Spring, 2011 - 30 Bite-Sized Tips for Best vSphere and Hyper-V...Concentrated Technology
The document provides 30 tips for optimizing virtual machine performance. Some key tips include purchasing hardware compatible with virtualization, using paravirtualized drivers for networking and storage, properly allocating CPU and memory resources to VMs, avoiding overuse of snapshots, performing resource-intensive tasks during off-hours, enabling jumbo frames and NTP time synchronization, and leveraging tools like DRS that prioritize faster hosts. Regular optimization and monitoring of VM configurations and underlying hardware is emphasized for maintaining good performance.
VMworld 2014: Virtualize Active Directory, the Right Way!VMworld
Virtualizing Active Directory domain controllers can provide benefits like increased availability and scalability. However, there are some safety considerations to take into account, such as preventing "USN rollback" which occurs when a domain controller's state is reverted, like after restoring from a snapshot. New features in Windows Server 2012 and VMware vSphere help address this, such as the VM Generation ID which changes when the domain controller state is modified, triggering safety mechanisms to isolate changes. Proper configuration following best practices is important for successfully virtualizing Active Directory.
Haiku Deck is a presentation platform that allows users to create Haiku-style slideshows. The document encourages the reader to get started creating their own Haiku Deck presentation on SlideShare by providing a link to do so. It aims to inspire the reader to try out Haiku Deck's unique presentation style.
Stories of climate change in southern Shan StateLIFT Fund
Watch this presentation to learn more about how LIFT is helping farmers in southern Shan State deal with the effects of climate change.
Visit www.lift-fund.org for more on LIFT.
Cassandra Summit 2015: Real World DTCS For OperatorsJeff Jirsa
Real World DTCS For Operators
The introduction of DateTieredCompactionStrategy in late 2014 was a significant step forward in providing a viable compaction strategy for time series data, especially time series data that will be TTL'd out. DateTieredCompactionStrategy's introduction was met with genuine excitement, and its rapid adoption is testament to developers' and operators' desire to have data compacted in a way that better matches their write patterns.
However, DateTieredCompactionStrategy's features come with significant limitations. This talk will review our real world benchmarking and use cases for DTCS as a vehicle to discuss the implications of DateTieredCompactionStrategy on operational tasks such as repair, read-repair, bootstrapping, and especially DR recovery scenarios, and it will also discuss how those various limitations lead us to proposing an operations-friendly alternative to DateTieredCompactionStrategy.
A day in the life of a VSAN I/O - STO7875Duncan Epping
This document provides an overview and summary of a VMworld session about Virtual SAN I/O. The session covers Virtual SAN concepts, the I/O flow of reads and writes in Virtual SAN, failure scenarios and how Virtual SAN handles them, and new features like deduplication and compression. The document includes diagrams demonstrating how data is distributed and replicated across hosts in a Virtual SAN cluster. It also provides details on how reads, writes, and failures are handled at a technical level in Virtual SAN. In the conclusion, it recommends three ways for attendees to get started with Virtual SAN: a hands-on lab, 60-day free evaluation, or working with a VMware partner on an assessment.
Virtualizing Tier One Applications - VarrowAndrew Miller
This document provides best practices for virtualizing mission critical applications like Exchange and SQL Server. It discusses the top 10 myths about virtualizing business critical applications and provides the truths. It then discusses best practices for virtualizing Exchange, including starting simple, licensing, storage configuration, and high availability options. For SQL Server, it covers starting simple, licensing, storage configuration, migrating, and database best practices. It also discusses tools that can be used for database performance analysis when virtualized like Confio IgniteVM and vCenter Operations.
- The document provides best practices for deploying and managing VMware vSphere environments. It covers topics such as ESXi host deployment, virtual machine deployment, vSphere HA clusters, vSphere networking, snapshots, security, vCenter deployment, backups, technical support, and housekeeping.
- Key recommendations include using compatible hardware, stable versions, host profiles, latest virtual hardware, syspreped templates, DRS automation, separate infrastructure and VM traffic, vSphere distributed switches, meaningful snapshots, external databases for vCenter, backups, and change management.
The have no fear guide to virtualizing databasesSolarWinds
When it comes to a successful database virtualization journey, there are things you must know before you start. In this presentation you will:
-Review terms and concepts for VMware, by far the most common virtualization platform
-Examine how to use vSphere (the VMware admin console)
-Explore the differences between virtual and physical host metrics a
-Learn to overcome the shortcomings of virtualizing your database environment
vSphere APIs for performance monitoringAlan Renouf
The document discusses vSphere APIs for performance monitoring and provides an overview of useful performance metrics such as CPU, memory, disk, and network. It motivates the need for these APIs by demonstrating how charts and visualizations can help identify performance issues. The agenda covers common useful statistics, how vSphere retrieves performance data, and how to access the data programmatically using tools like PowerCLI and Java/C#.
Abstract:
Reactive applications need to be able to respond to demand, be elastic and ready to scale up, down, in and out—taking full advantage of mobile, multi-core and cloud computing architectures—in real time.
In this talk we will discuss the guiding principles making this possible through the use of share-nothing and non-blocking designs, applied all the way down the stack. We will learn how to deliver systems that provide reactive supply to changing demand.
I gave this talk at React Conf 2014 in London. Recording available here: https://www.youtube.com/watch?v=mBFdj7w4aFA
Presentation architecting a cloud infrastructurexKinAnx
This document provides an agenda and overview for a session on architecting a cloud infrastructure. The agenda includes introductions, gathering requirements, sizing and scaling, host design, vCenter design, cluster design, networking and storage considerations. It emphasizes the importance of gathering requirements from customers and conceptualizing the design based on those requirements. It also discusses various design considerations and best practices for each component of a cloud infrastructure.
SaltStack can be used to automate and orchestrate the provisioning of virtual machines on VMware ESXi 6.0. It implements the VMware APIs to allow defining VM profiles and templates that specify VM configurations, and then uses Salt commands to rapidly deploy new VMs from templates with customized configurations. Open-VM tools must be installed on templates to enable customizing VMs, such as setting the network configuration. Salt files define VM profiles and provider credentials, separating configuration from deployment logic for flexibility and reusability.
How we collaborated with the CentOS and Xen projects to build a next-generation platform at Go Daddy. Discussion of the design considerations, infrastructure, succes stories and challenges of this paradigm change
This document is a slide deck presentation about high availability (HA) and distributed resource scheduler (DRS) gotchas that can negatively impact virtual infrastructure. It discusses common issues administrators encounter with vMotion capabilities as hardware is refreshed over time. It also explains how DRS and HA work together to optimize resource utilization and provide fault tolerance through automated workload balancing and live migration of virtual machines during host failures. The presentation emphasizes properly configuring clusters, resource pools, and rules to allow DRS and HA to fully leverage vMotion's capabilities.
VMworld 2011 Review: Preparing for vSphere 5 with Virtualization ManagerSolarWinds
For more information on Virtualization Manager, visit: http://www.solarwinds.com/virtualization-manager.aspx
Watch this webcast: http://www.solarwinds.com/resources/webcasts/vmworld-us-takeaways-insights-and-best-practices.html
During this Geek Speak webcast, you will learn why Virtualization Manager won the Best of VMworld 2011 Gold Award for Virtualization Management and how to solve the issues affecting you the most.
Slides from the second meeting of the Toronto High Scalability Meetup @ http://www.meetup.com/toronto-high-scalability/
-Basics of High Scalability and High Availability
-Using a CDN to Achieve 99% Offload
-Caching at the Code Layer
Virtualization allows computer resources to be abstracted and shared. It involves simulating hardware to run virtual machines that behave like physical computers. A hypervisor manages virtual machines and allocates resources to them. Virtualization provides benefits like efficient resource use, increased availability, disaster recovery, and on-demand scaling. While it adds complexity, virtualization is commonly used to virtualize desktops, run specific programs, create test environments, design private clouds, and utilize public clouds. Major vendors provide virtualization software and cloud services.
24 Hours of PASS, Summit Preview Session: Virtual SQL Server CPUsDavid Klee
This document discusses virtual CPUs and CPU architecture. It begins by explaining how hypervisor resource queues work and how requests for CPU, memory, storage and networking are placed in queues. It then covers physical CPU architecture including cores, sockets, NUMA and memory locality. It discusses how virtual CPUs are scheduled by the hypervisor and ways to measure scheduling pressure. Finally, it provides recommendations for right-sizing virtual machines and balancing workloads to reduce scheduling delays.
This document provides an overview of implementing affordable disaster recovery with Hyper-V and multi-site clustering. It discusses what constitutes a disaster, the key components needed which are a storage mechanism, replication mechanism, and target servers/cluster. It also covers clustering history, what a cluster is, and the important concept of quorum which determines a cluster's existence through voting of its members.
The popularity of Virtual SAN is growing daily. Server admins are finally free to aggregate storage in their servers to create a shared storage system that scales with their compute needs. The underlying key to making it all work is networking. All Virtual SAN data flows through it, and correct selection and configuration of networking components will mean the difference between disruptive success or dramatic failure. This session will give deep insight in the do's and don'ts of Virtual SAN networking. Best practices for physical and virtual switch configuration and performance testing will be discussed. Virtual SAN 5.5 and 6.0 will be covered, and the networking differences discussed. Methods of troubleshooting network issues will be covered. For those configuring a Virtual SAN network for the first time, for labs or enterprise scale, this session is a must-see.
WinConnections Spring, 2011 - 30 Bite-Sized Tips for Best vSphere and Hyper-V...Concentrated Technology
The document provides 30 tips for optimizing virtual machine performance. Some key tips include purchasing hardware compatible with virtualization, using paravirtualized drivers for networking and storage, properly allocating CPU and memory resources to VMs, avoiding overuse of snapshots, performing resource-intensive tasks during off-hours, enabling jumbo frames and NTP time synchronization, and leveraging tools like DRS that prioritize faster hosts. Regular optimization and monitoring of VM configurations and underlying hardware is emphasized for maintaining good performance.
VMworld 2014: Virtualize Active Directory, the Right Way!VMworld
Virtualizing Active Directory domain controllers can provide benefits like increased availability and scalability. However, there are some safety considerations to take into account, such as preventing "USN rollback" which occurs when a domain controller's state is reverted, like after restoring from a snapshot. New features in Windows Server 2012 and VMware vSphere help address this, such as the VM Generation ID which changes when the domain controller state is modified, triggering safety mechanisms to isolate changes. Proper configuration following best practices is important for successfully virtualizing Active Directory.
Haiku Deck is a presentation platform that allows users to create Haiku-style slideshows. The document encourages the reader to get started creating their own Haiku Deck presentation on SlideShare by providing a link to do so. It aims to inspire the reader to try out Haiku Deck's unique presentation style.
Stories of climate change in southern Shan StateLIFT Fund
Watch this presentation to learn more about how LIFT is helping farmers in southern Shan State deal with the effects of climate change.
Visit www.lift-fund.org for more on LIFT.
Patient XW has a history of osteoporosis and fractures. Bone density scans showed her condition has worsened, with T-scores below -2.5. She was initially treated with bisphosphonates but had adverse effects. The treatment plan was changed to denosumab (Prolia) injections every 6 months, as it is effective for patients who do not respond to or cannot tolerate bisphosphonates. Her insurance may require prior authorization for the costly biologic treatment.
Mr. Mwalimu Raphael Jilani has extensive experience in microbiology. He received his MSc in Microbiology from University College of Science, Osmania University in 2015 and BSc in Microbiology, Genetics and Chemistry from Nizam College, Osmania University in 2013. He has worked with Bola Associates Ltd since 2016 marketing water cleaning services and advising on water technologies. Prior to this he had an internship at the Government Chemist Department in Mombasa performing laboratory analysis and an attachment at Coast General Hospital processing clinical specimens and performing antibiotic testing. He also taught mathematics and chemistry at Paul Harris High School in 2010.
This document provides an overview of iintegra, a recruitment software solution. It begins with the evolving recruitment landscape and challenges businesses face. It then outlines iintegra's recruitment process, including simplifying the process through automation. The document demonstrates iintegra's features like applicant tracking and reporting. It concludes with an overview of iintegra's services, setup process, and potential benefits and ROI for customers.
1) This section provides background on sustainable development and political leadership by reviewing relevant literature. Sustainable development aims to meet current needs without compromising future generations' ability to meet their own needs.
2) It discusses how sustainable development focuses on maintaining a balance between development and the environment. Sustainable development encompasses ecology, economics, politics and culture.
3) The review informs the research question by exploring concepts of political leadership and sustainable urban development. It helps shape the conceptual framework and hypotheses around factors that motivate local leaders to implement sustainable development policies.
Este documento describe la nanotecnología y sus aplicaciones. Explica que la nanotecnología involucra el desarrollo de dispositivos extremadamente pequeños que pueden tener nuevas propiedades físicas, químicas y biológicas. Luego menciona algunas aplicaciones como nanobolsas purificadoras de agua, nanopartículas protectoras para circuitos y nanoesponjas protectoras para el cuerpo. Finalmente, brinda información biográfica básica sobre Richard Feynman.
Haiku Deck is a presentation platform that allows users to create Haiku-style slideshows. The document encourages the reader to get started creating their own Haiku Deck presentation on SlideShare by providing a link to do so. It aims to inspire the reader to try out Haiku Deck's unique presentation style.
CSOFT International product TermWiki Pro, an innovative cloud based terminology management system. Capable of meeting all of the manufacturer’s requirements, it is the ideal tool for terminology creation, management, and dissemination.
This short document promotes the creation of presentations using Haiku Deck, an online presentation tool. It includes two stock photos and text suggesting the reader may be inspired to create their own Haiku Deck presentation. A call to action is given to get started making a presentation on SlideShare.
A presentation on time management. Used as the basis for a video lecture in the course, "Creating a Sustainable Writing Process," offered by Eastlake & Roanoke.
"Why you should rent a Gagal Home serviced apartmentGagalhome
Gagal Home & Astha Hospitality offers Fully Furnished cost effective Service Apartment in Mumbai, at best price. Our service apartments located at Andheri, Kandivali, Goregaon, Borivali & Bandra Kurla Complex (BKC).
This document summarizes test results for the PULSE EHR product version 14.2.3 developed by Community Health Systems. The testing was conducted by SLI Global Solutions and certified by ICSA Labs. The product was tested for the inpatient setting and certified for multiple 2014 Edition criteria both directly and through inheritance. It relies on Medicity and MedHost software. Testing validated usability of clinical decision support functionality.
This document discusses sustainability projects including the repaving of a perimeter road before a rainstorm, one of Henry Segerstrom's last gifts of Macy's South Coast Plaza in 2015, the expanding and contracting tides of the Newport Coast Estuary and cautionary environments, surf increasing in Huntington Beach in 2015 possibly due to climate, wind turbines being leaders in renewable energy, architecture and design projects, wineries, airports, utility-scale grid upgrades, and sky noon.
Mridul Sharma is an experienced Assistant Manager with over 9 years of experience in call center operations and training at Vodafone Mobile Services Ltd. He possesses excellent communication, leadership, and customer service skills. He has received numerous awards and recognition for his work launching projects, maximizing sales, and completing trainings. His objective is to work for a company with opportunities for growth where his experience can be leveraged for mutual benefit.
Mike James will speak on Sunday May 1st from 6-7:30pm about his book. There will also be an event in Frankfort on May 4th at noon for information and signups. A 6 week outreach program called Reaching Richmond begins April 28th at 6:30pm led by Gary Maynard, sign up in the FLC. People are asked to bring framed photos of mothers or with children by May 6th for display on Mother's Day.
VMworld 2013: DRS: New Features, Best Practices and Future Directions VMworld
The document discusses new features and future directions for VMware Distributed Resource Scheduler (DRS). Key points include:
1) DRS 5.5 introduces features like automatically tuning the number of VMs per host and better handling of latency-sensitive and CPU-intensive workloads.
2) DRS is integrated with new storage technologies like VMware vFlash and vSAN. It also supports autoscaling of proxy switch ports.
3) Future areas of focus include network DRS with bandwidth reservations, more accurate static VM overhead memory estimation, and proactive DRS monitoring for potential issues.
VMworld 2013: Operating and Architecting a vSphere Metro Storage Cluster base...VMworld
VMworld 2013
Lee Dilworth, VMware
Learn more about VMworld and register at http://www.vmworld.com/index.jspa?src=socmed-vmworld-slideshare
Duncan Epping, VMware
The document provides recommendations for optimizing CPU performance in virtualized environments:
1. Generally, 4-6 vCPUs per physical socket is recommended, as hyperthreading does not increase processing capacity.
2. Reserve CPUs for high-performance VMs to ensure they receive required CPU cycles.
3. Reduce vCPUs assigned to VMs if guest CPU usage is low to optimize performance.
4. Monitor CPU usage and ready times for VMs and hosts to identify potential overload issues. Right-sizing VMs, adjusting overcommitment levels, and ensuring balanced workloads can help address high CPU ready times with low utilization.
The document outlines 7 lessons for building highly effective real-time servers from DK Moon's experience working on game servers. The lessons are: 1) Know your traffic pattern to minimize network usage. 2) Avoid packet copying to reduce CPU usage. 3) Focus on client obfuscation over encryption. 4) Use multicasting instead of broadcasting. 5) Don't rely on databases for synchronization. 6) Caching introduces overhead so use directly when possible. 7) Develop monitoring tools before launching to ensure visibility.
Master VMware Performance and Capacity ManagementIwan Rahabok
12 Sep 2016 update: See this http://virtual-red-dot.info/operationalize-sddc-program-2/ for details.
-------------
Based on the book http://virtual-red-dot.info/performance-and-capacity-management/
Master performance and capacity management of VMware SDDC
The document provides 30 tips for optimizing virtual machine performance. Some key tips include allocating only the CPUs and RAM needed for each VM, upgrading to VMware tools version 7 for performance improvements, enabling hyperthreading, setting NICs to autonegotiate, and disconnecting unused physical hardware devices from VMs. The document emphasizes the importance of right-sizing resources for each VM based on its specific workload.
This document is a slide deck about Hyper-V high availability and live migration presented by Greg Shields of Concentrated Technology. The deck covers understanding live migration and its role in Hyper-V HA, fundamentals of Windows failover clustering, building a two-node Hyper-V cluster with iSCSI storage, managing a Hyper-V cluster, and adding disaster recovery with multi-site clustering. The deck is intended to help IT professionals implement and manage highly available Hyper-V environments.
This document summarizes notes from the Amazon Builders Library about designing distributed systems. It discusses techniques like avoiding one-way decisions, ensuring rollback safety during deployments, using local and external caches effectively, implementing health checkers, load shedding to avoid overload, avoiding queue backlogs, and instrumenting systems for observability. The Amazon Builders Library contains articles on best practices for building AWS applications that are performant, robust, and reliable.
Webinar slides: 9 DevOps Tips for Going in Production with Galera Cluster for...Severalnines
This document provides an overview and agenda for a webinar on managing and monitoring MySQL clusters using ClusterControl by Severalnines. The webinar host is introduced and instructions are provided for asking questions. The webinar will cover topics such as operating system configuration, backup strategies, replication, query performance, schema changes, security, reporting, and disaster recovery. Case studies and customers are also briefly mentioned.
Video and slides synchronized, mp3 and slide download available at URL https://bit.ly/2BubBkw.
Alex Robinson walks through his experiences trying to reliably run a distributed database on Kubernetes, optimize its performance, and help others do the same in their heterogeneous environments. He looks at what kinds of stateful applications can most easily be run in containers, and a number of pitfalls he encountered along the way. Filmed at qconsf.com.
Alex Robinson is a member of the technical staff at Cockroach Labs, where he works on CockroachDB's core transactional storage layer and leads all integrations with orchestration systems. Previously, he was a senior software engineer at Google, where he spent his last two years as a core early developer of Kubernetes and GKE.
Accelerate with ibm storage ibm spectrum virtualize hyper swap deep divexKinAnx
The document provides an overview of IBM Spectrum Virtualize HyperSwap functionality. HyperSwap allows host I/O to continue accessing volumes across two sites without interruption if one site fails. It uses synchronous remote copy between two I/O groups to make volumes accessible across both groups. The document outlines the steps to configure a HyperSwap configuration, including naming sites, assigning nodes and hosts to sites, and defining the topology.
This document provides an overview and summary of key points from a presentation on designing virtual infrastructures and hypervisors. It discusses pre-requisites, assessing which servers are good candidates for virtualization, measuring server performance, determining the right amount of RAM for virtual machines, different types of virtualization technologies, high availability options, and live migration capabilities.
Linux High Availability provides information on configuring high availability clusters in Linux. It discusses:
- Key components of HA clustering including one service taking over work of another if it fails, IP and service takeover.
- The importance of high availability and costs of downtime for businesses. Statistics show significant downtime even at 99.9% availability levels.
- Best practices for HA including keeping configurations simple, preparing for failures, and testing HA setups. Complexity increases reliability risks.
Presentation architecting a cloud infrastructuresolarisyourep
This document provides an agenda and overview for a session on architecting a cloud infrastructure. The agenda includes introductions, gathering requirements, sizing and scaling, host design, vCenter design, cluster design, networking and storage considerations. It emphasizes the importance of gathering requirements from customers and conceptualizing the design based on those requirements. It also discusses various design considerations and best practices for each component of a cloud infrastructure.
Presentation drs advanced concepts, best practices and future directionssolarisyourep
DRS provides advanced resource management capabilities for VMware vSphere environments. It uses resource pools and controls to optimize VM placement and workload balancing across hosts. The primary goals of DRS are to maintain high VM performance by meeting their resource demands and keeping applications happy. While load balancing is also important, it is secondary to ensuring VMs receive sufficient resources. DRS includes features for handling complex scenarios with multiple constraints and metrics. Future enhancements may include more aggressive load balancing modes and automatic tuning of configuration parameters.
Retaining Goodput with Query Rate LimitingScyllaDB
ScyllaDB uses a shared-nothing architecture where data is split into partitions across nodes and shards. The "hot partition problem" can occur when a partition becomes overloaded, impacting other nearby partitions. To address this, ScyllaDB implements per-partition rate limiting which counts operations and rejects some to keep the rate under a defined limit. Exceptions were initially making rejections expensive, but this was addressed by avoiding exceptions or implementing missing exception inspection capabilities. In benchmarks, rate limiting restored goodput and provided more stable performance under timeouts.
VMworld Europe 2014: Virtual SAN Best Practices and Use CasesVMworld
This document provides an overview and agenda for a presentation on VMware Virtual SAN. It discusses key features of Virtual SAN including its software-defined storage approach and hybrid storage using SSD and HDD. Several use cases are reviewed like virtual desktop infrastructure, remote office/branch office, and DMZ/isolated environments. Best practices are also covered for various use cases around sizing, policies, and ready nodes. The document aims to introduce attendees to Virtual SAN capabilities and considerations for different deployment scenarios.
This document provides an overview of a training course on Microsoft's Hyper-V virtualization platform. The course is divided into two halves with breaks in between sections. The first half covers introductory topics like Hyper-V infrastructure, networking, storage and management. The second half focuses on more advanced topics such as high availability, integration with System Center, and other Microsoft virtualization technologies. The document also includes several pages on Hyper-V concepts like NUMA support, dynamic memory, and differences between the Hyper-V Server and Windows Server editions.
Retaining Goodput with Query Rate LimitingScyllaDB
Distributed systems are usually optimized with particular workloads in mind. At the same time, the system should still behave in a sane way when the assumptions about workload do not hold - notably, one user shouldn't be able to ruin the whole system's performance. Buggy parts of the system can be a source of the overload as well, so it is worth considering overload protection on a per-component basis. For example, ScyllaDB's shared-nothing architecture gives it great scalability, but at the same time makes it prone to a "hot partition" problem: a single partition accessed with disproportionate frequency can ruin performance for other requests handled by the same shards. This talk will describe how we implemented rate limiting on a per-partition basis which reduces the performance impact in such a case, and how we reduced the CPU cost of handling failed requests such as timeouts (spoiler: it's about C++ exceptions).
Similar to Presentation avoiding the 19 biggest ha & drs configuration mistakes (20)
VMware is a market leader in virtualization software with over $3.7 billion in revenues and over 13,000 employees worldwide. The document discusses how VMware's technologies enable a new era of IT by providing a more flexible, scalable and efficient infrastructure that supports applications across existing datacenters and public cloud services. This allows for an empowered and secure mobile workforce with faster time-to-market for modern applications.
Presentation a vision for user centric computingsolarisyourep
This document summarizes a presentation about transitioning from a PC-centric to a user-centric computing model using VMware View virtual desktop infrastructure. It discusses how VMware View provides benefits to both IT and end users by enabling flexible access to desktops and applications from any device while giving IT control and reducing costs. It also provides an overview of VMware View components and capabilities for cloud infrastructure, IT control, and end user freedom.
Presentation advanced management – the road aheadsolarisyourep
The document discusses VMware's approach to management and automation for IT organizations. It introduces several new management suites from VMware that aim to help IT operate more like a business and deliver value to the business. The suites discussed are the vCenter Operations Management Suite, vFabric Application Management Suite, and IT Business Management Suite. The suites are designed to simplify management, increase automation, provide visibility across infrastructure and applications, and help IT articulate its value using business metrics and language.
Presentation architecting virtualized infrastructure for big datasolarisyourep
The document discusses how virtualization can help simplify big data infrastructure and analytics. Key points include:
1) Virtualization can help simplify big data infrastructure by providing a unified analytics cloud platform that allows different data frameworks and workloads to easily share resources.
2) Hadoop performance on virtualization has been proven with studies showing little performance overhead from virtualization.
3) A unified analytics cloud platform using virtualization can provide benefits like better utilization, faster provisioning of elastic resources, and multi-tenancy for secure isolation of analytics workloads.
Presentation blade center foundation for cloudsolarisyourep
This document provides an overview and summary of IBM BladeCenter Foundation for Cloud. It discusses the current environment and need for IT to help address business challenges through virtualization and cloud computing. It then introduces IBM BladeCenter Foundation for Cloud as an integrated virtualized platform that can be deployed quickly to provide a foundation for cloud computing. Key features highlighted include built-in server virtualization, startup services, hardware configurations ranging from small to extra large, and major hardware components.
Presentation building and running your private cloudsolarisyourep
This document discusses how IT can become an internal cloud service provider by using VMware's vCloud solutions. It describes how IT can standardize services, organize users into organizations, and provide self-service provisioning of virtual resources through catalogs. Users are given automated and rapid access to resources through a pay-as-you-go model while IT maintains control through chargeback, user limits, approvals, and access controls. The document also recommends VMware consulting services to help guide the transformation to a private cloud environment.
Presentation building your cloud with v mwaresolarisyourep
This document discusses networking models in VMware vCloud. It explains that there are three layers of networking: external, organization, and vApp networks. External networks are outside of vCloud and provide connectivity to organization networks. Organization networks allow communication within an organization and can be connected to external networks. vApp networks are contained within a vApp. The document provides details on how each network type is configured and managed.
Presentation business critical applications in a virtual envsolarisyourep
This document discusses virtualizing business critical applications. It begins by explaining why IT operations and application owners sometimes differ on virtualization, with the former wanting infrastructure efficiency and the latter concerned about performance and support. The document then shows trends of increasing virtualization for various workloads. It outlines operational benefits and new features in vSphere 5 that further reduce barriers to 100% virtualization. The rest of the document focuses on how virtualization can improve quality of service, availability, and time-to-market for applications through features like dynamic resource allocation, high availability, disaster recovery, and faster test/development cycles. It also addresses specific application information and licensing models. In the end it provides recommendations for getting started with virtualizing critical applications.
Presentation cim1309 v cat 3.0 operating a v-mware cloudsolarisyourep
This document discusses the VMware vCloud Architecture Toolkit (vCAT) 3.0 and focuses on the "Operating a VMware Cloud" section. It provides an overview of cloud operations and frameworks for process maturity. It also covers organizing cloud operations, including roles and teams. Finally, it describes key areas for cloud business and consumer control, service control, operations control, and infrastructure control. The goal is to help organizations understand how to operate a cloud based on VMware technologies and best practices.
Presentation cisco intelligent automation complementing and extending v mwa...solarisyourep
This document discusses how Cisco Intelligent Automation can complement and extend VMware vCloud Director. It notes that vCloud Director provides standardization, increased agility and application portability through compute models. Cisco Intelligent Automation adds value-added services, simplified configuration and intelligent automation. It can orchestrate all datacenter resources required to support vCloud, provide a unified portal for virtual, cloud, physical and professional services, and deliver additional services like monitoring, backup and disaster recovery for vApps.
Presentation cisco vxi–optimized infrastructure for scaling v mware view wi...solarisyourep
This document provides an overview and summary of a Cisco presentation on optimizing infrastructure for scaling VMware View desktop virtualization. The key points discussed include:
- Rising desktop management costs and the drivers for desktop virtualization like mobility, security, and flexibility.
- Cisco's Virtualization Experience Infrastructure (VXI) which provides validated designs, security, management, and an ecosystem of partners to deliver a superior virtual desktop experience.
- How the Cisco Unified Computing System (UCS) provides an ideal architecture for desktop virtualization through scalability, density, visibility and integrated networking.
Presentation cloud infrastructure and management – from v sphere to vcloud ...solarisyourep
- VMware is a market leader in virtualization and cloud infrastructure software, with over $3.77 billion in revenues in 2011. It has over 300,000 customers including all of the Fortune 100 companies.
- The document discusses VMware's product portfolio including vSphere, vCloud Director, vShield, and vCenter which provide capabilities for virtualization, private clouds, security and management. It also discusses how these products work together to enable hybrid cloud environments.
- Case studies are presented showing how NYSE Euronext and SAP use VMware's virtualization and cloud solutions to improve the flexibility, availability and cost-efficiency of their IT infrastructure and applications.
VMware announced new versions of its cloud infrastructure software including vSphere 5, vCenter 5, vCloud Director 1.5, and vCenter Site Recovery Manager 5. The updates focus on accelerating the path to 100% virtualization, providing granular control of network and storage resources, and delivering an intelligent virtual infrastructure across private and public clouds.
The document discusses how cloud computing and big data are impacting virtualized environments. It summarizes that cloud architecture is well-suited for hosting mission critical applications, while big data requires new scalable infrastructure like scale-out storage. VMware provides a platform that can seamlessly support both cloud and big data workloads, allowing for mobility and control across applications. EMC offers solutions powered by Intel technologies that leverage these trends, including virtual storage, data replication, and scale-out storage for big data.
The document discusses consuming cloud resources from the perspective of a consumer. It introduces vCloud Architecture Toolkit (vCAT) as a vendor-agnostic set of design guidelines and best practices for architects. It then outlines different ways consumers can access cloud resources, including through vFabric Application Director, a custom portal using vCloud Director SDK or PowerCLI, or vCenter Orchestrator. The document is presented as part of a discussion on consuming cloud resources.
Presentation desktops for the cloud the view rolloutsolarisyourep
This document provides an overview of VMware View and its implementation at a company. View allows desktop operating systems to run virtualized in a data center rather than locally on devices. It discusses the benefits of View like easier maintenance, automatic backups, and faster logins. It also outlines the company's rollout methodology, including determining user profiles, finalizing app lists, and agreeing on a timeline. The typical virtual machine configuration is also specified.
Presentation disaster recovery in virtualization and cloudsolarisyourep
This document discusses business continuity and disaster recovery strategies in virtual and cloud environments. It outlines different types of availability designs including stretched clusters across sites, multiple vSphere clusters, and site-to-site replication with disaster recovery. It explains when to use stretched vSphere clusters versus site recovery manager, and discusses features of vSphere replication. The document aims to help customers understand different options and select the right solution based on their requirements for disaster avoidance, recovery, and planned migration.
Presentation end-user computing in the post-pc erasolarisyourep
VMware provides a platform to help organizations transition to the post-PC era by connecting, managing, and simplifying access to users, desktops, applications, and data across any device. The platform includes View for virtual desktops, Horizon for mobile workspaces, Project AppBlast for universal application access, and Project Octopus for secure data services. This allows IT to accelerate the transition while delivering delighted users through a common managed infrastructure.
Presentation expect more from your storagesolarisyourep
The document discusses IBM storage solutions that provide essential storage efficiency technologies, storage systems that analyze and adapt to improve performance, and cloud-ready storage features. It promotes IBM's Storwize V7000 unified storage system with integrated file and block storage, its XIV Gen3 self-tuning virtualized storage, and Easy Tier and Quality of Service Management for auto-tiering and performance guarantees. The document argues that these IBM storage solutions allow customers to expect more from their storage.
Presentation from sap virtualization to hybrid cloudssolarisyourep
VMware is the leading virtualization platform for SAP workloads. Case studies show virtualizing SAP with VMware can significantly reduce costs while improving provisioning times. VMware's hybrid cloud strategy allows organizations to extend their existing datacenters into public clouds, providing a common platform and security across private and public resources.
Presentation from sap virtualization to hybrid clouds
Presentation avoiding the 19 biggest ha & drs configuration mistakes
1. Avoiding the 19 Biggest
HA & DRS Configuration
Mistakes - 2012 Edition
Greg Shields, Concentrated Technology
INF-VSP1232
#vmworldinf
2. Who is that Ponytailed Guy?
• Greg Shields, MVP, vExpert, CTP
● Senior Partner, Concentrated Technology
● www.ConcentratedTech.com
● @ConcentratdGreg
• Over 15 years of IT experience.
● Consultant – SMB to Enterprise…
● Speaker – TechMentor, Tech Ed, Windows Connections,
MMS, VMworld, countless others…
● Author – Sixteen books and counting…
● Columnist – TechNet Magazine, Redmond Magazine,
Windows IT Pro Magazine, others…
5. Reality Moment: HA/DRS Solve Two Problems
Problem #1:
Protection from
Unplanned Host
Downtime
6. Reality Moment: HA/DRS Solve Two Problems
Problem #2:
Load Balancing and
Defragmentation of
Resources
7. Contrary to Popular Belief…
• …seeing a vMotion occur isn’t all that exciting.
8. Contrary to Popular Belief…
• …seeing a vMotion occur isn’t all that exciting.
• DEMO: Watching a
vMotion occur…
9. Useful, However…
• …is recognizing where bad HA and DRS settings
impact vMotion’s ability to do its job.
● A surprising number of environments have configured HA/DRS
settings incorrectly.
● Some do so because of hardware constraints.
● Others have not designed architecture with HA/DRS in mind.
● Even others have introduced problems as they scale upwards.
• What follows are 16 19 big mistakes you’ll want to avoid
as you build or scale your HA/DRS cluster(s).
10. Big Mistake #1:
Not Planning for HW Evolution
Successful vMotion requires similar processors.
● Processors must be from the same manufacturer.
• No Intel-to-AMD or AMD-to-Intel vMotioning.
● Processors must be of a proximate families.
● This bites people a-few-years-down-the-road all the time!
12. Big Mistake #1:
Not Planning for HW Evolution
As a virtual environment ages, hardware is refreshed
and new hardware is added.
● Refreshes sometimes create “islands” of vMotion capability
How can we always vMotion between computers?
13. Big Mistake #1:
Not Planning for HW Evolution
As a virtual environment ages, hardware is refreshed
and new hardware is added.
● Refreshes sometimes create “islands” of vMotion capability
How can we always vMotion between computers?
● You can always refresh all hardware at the same time (Har!)
14. Big Mistake #1:
Not Planning for HW Evolution
As a virtual environment ages, hardware is refreshed
and new hardware is added.
● Refreshes sometimes create “islands” of vMotion capability
How can we always vMotion between computers?
● You can always refresh all hardware at the same time (Har!)
● You can cold migrate, with the machine powered down.
(Always works, but ain’t all that friendly.)
15. Big Mistake #1:
Not Planning for HW Evolution
As a virtual environment ages, hardware is refreshed
and new hardware is added.
● Refreshes sometimes create “islands” of vMotion capability
How can we always vMotion between computers?
● You can always refresh all hardware at the same time (Har!)
● You can cold migrate, with the machine powered down.
(Always works, but ain’t all that friendly.)
● You can use vMotion Enhanced Compatibility Mode to manage
your vMotion-ability. Create islands as individual clusters.
SOLUTION: vMotion EVC
17. Big Mistake #2:
Not Planning for svMotion
• Storage vMotion has some special requirements.
● Virtual machines with snapshots cannot be svMotioned.
● Virtual machine disks must be persistent mode or RDMs.
● The host must have sufficient resources to support two instances
of the VM running concurrently for a brief time.
● The host must have a vMotion license, and be correctly
configured for vMotion.
● The host must have access to both the source and target
datastores.
18. Big Mistake #3:
Not Enough Cluster Hosts
• You cannot change the laws of physics.
● For HA to failover a VM, there must be resources available
elsewhere in the cluster.
● These resources must be set aside. Reserved. “Wasted”.
19. Big Mistake #3:
Not Enough Cluster Hosts
• You cannot change the laws of physics.
● For HA to failover a VM, there must be resources available
elsewhere in the cluster.
● These resources must be set aside. Reserved. “Wasted”.
• Many environments don’t plan for
cluster reserve when designing
their clusters.
● Nowhere for VMs to go…
20. Big Mistake #3:
Not Enough Cluster Hosts
• A fully-prepared cluster must set aside one full server’s
worth of resources in preparation for HA.
21. Big Mistake #3:
Not Enough Cluster Hosts
• A fully-prepared cluster must set aside one full server’s
worth of resources in preparation for HA.
● This is done in your Admission Control Policy.
● First, Enable Admission Control.
22. Big Mistake #3:
Not Enough Cluster Hosts
• A fully-prepared cluster must set aside one full server’s
worth of resources in preparation for HA.
● This is done in your Admission Control Policy.
● Then, set Host failures cluster tolerates to 1 (or more).
23. Big Mistake #3:
Not Enough Cluster Hosts
• A fully-prepared cluster must set aside one full server’s
worth of resources in preparation for HA.
● This is done in your Admission Control Policy.
● Then, set Host failures the cluster tolerates to 1 (or more).
THE
24. Big Mistake #4:
Setting Host Failures the Cluster Tolerates to 1.
• Setting Host failures the cluster tolerates to 1 may
unnecessarily set aside too many resources.
● Not all your VMs are Priority One.
● Some VMs can stay down if a host dies.
● Setting aside a full host is wasteful,
particularly when your number of hosts
is small.
25. Big Mistake #4:
Setting Host Failures the Cluster Tolerates to 1.
• Tune your level of waste with Percentage of cluster
resources reserved as failover capacity.
● Set this to a lower value than one server’s contribution.
26. Big Mistake #5:
Forgetting to Prioritize VM Restart.
• VM Restart Priority is one of those oft-forgotten settings.
● A default setting is configured when you enable HA.
● Per-VM settings must be configured for each VM.
• These settings are most important during an HA event.
● Come into play when Percentage policy is enabled.
• Easter Egg:
Restart Policy is Per-Host!
27. Big Mistake #6:
Disabling Admission Control
• Every cluster with HA enabled will have “waste”.
● Some enterprising young admins might enable HA but disable
Admission Control.
● “A-ha,” they might say, “This gives me all the benefits of HA but
without the waste!”
28. Big Mistake #6:
Disabling Admission Control
• Every cluster with HA enabled will have “waste”.
● Some enterprising young admins might enable HA but disable
Admission Control.
● “A-ha,” they might say, “This gives me all the benefits of HA but
without the waste!”
● They’re wrong. Squeezing VMs during an HA event can cause
downstream performance effects as hosts begin swapping.
● Never disable Admission Control.
29. Big Mistake #7:
Not Updating Percentage Policy
• The Percentage policy may need to be adjusted as your
cluster size changes.
● Adding servers changes the percentage of resources that must
be set aside.
● Take a look at adjusting percentage every time you add servers.
• Host failures the cluster tolerates needs no adjusting.
● No matter how many hosts you have, this policy setting will
always set aside one server’s worth of resources.
● Yet here danger lies…
30. Big Mistake #8:
Buying (the Occasional) Big
• Host failures the cluster tolerates sets aside the amount
of resources needed to protect every server.
● This means that any fully-loaded server will be HA-protected.
● Including, your biggest server!
• Thus, Host failures cluster tolerates must set aside
resources equal to your biggest server!
● If you buy three small servers and one big server, you’re wasting
even more resources!
31. Big Mistake #9:
Neglecting Host Isolation Response
• Host isolation response instructs the cluster what to do
when a host loses connectivity, but hasn’t failed.
● That host is isolated from the cluster, but its VMs still run.
32. Big Mistake #9:
Neglecting Host Isolation Response
• Host isolation response instructs the cluster what to do
when a host loses connectivity, but hasn’t failed.
● That host is isolated from the cluster, but its VMs still run.
• Three options available:
Leave Powered On / Power off / Shut down.
● VMs that remain powered on cannot be managed by surviving
cluster hosts. Egad, its like split brain in reverse!
● VMFS locks prevent them from being evacuated to “good” host.
● Shut Down will gracefully down a VM, but will release VMFS
locks so VM can be restarted elsewhere.
● Leave Powered On is useful, except where management and
VM networks are shared (e.g. Converged Networks).
33. Big Mistake #9:
Neglecting Host Isolation Response
• I Summon the Vast Power of Heartbeat Datastores!
● vSphere HA in v5.x can use the storage subsystem as a
secondary communication path.
● Adds redundancy.
● Used as communication
channel only when the
management network
is lost
● Such as in the case
of isolation or network
partitioning.
34. Big Mistake #10:
Assuming that Datastore Heartbeats
Prevent Isolation Events
• Datastore heartbeats are used by the master to
determine the state of an unresponsive host.
● They enable the master to determine whether the host
has failed completely…or is merely isolated.
• However: Isolation Response is triggered by the slave.
1. The slave is isolated (poor slave…)
2. The slave enters election state
3. The slave elects itself as master
4. The slave pings isolation addresses
5. The slave declares itself isolated and triggers isolation
response
35. Big Mistake #11:
Confusing your APD with your PDL
• An All Points Down scenario exists when all
communication is severed between host and device.
● I/O is then queued until a SCSI response code officially
reports the link is down.
● This can lead to indefinite queuing of device I/O
(and a very bad day for the affected VM).
• A Permanent Device Loss scenario exists when the host
can see the device target, but the target isn’t listening.
● Lets the host recognize that I/O should no longer be queued.
(E.g. “Give up, the device isn’t coming back.”)
36. Big Mistake #11:
Confusing your APD with your PDL
• APD is a more-common scenario in most environments,
but APD events will not trigger vSphere HA.
● The VM will run, but can’t read from or write to its disks.
• vSphere 5.0 U1 now allows HA to take action when a
datastore reaches a PDL state.
● Set the host setting disk.terminateVMOnPDLDefault to TRUE in
/etc/vmware/settings to ensure VMs are killed when their
datastore reaches a PDL state.
● Set the HA advanced setting das.maskCleanShutdownEnabled
to TRUE to trigger a restart response for any VM killed because
of the PDL condition.
● This is most handy for metro/stretch cluster configurations.
37. Big Mistake #12:
Overdoing Reservations, Limits, and Affinities
• HA may not consider these “soft affinities” at failover.
• However, they will be invoked after HA has taken its
corrective action.
● Reservations and limits can constrain resulting calculations.
● Affinities add more constraints, particularly in smaller clusters.
• Possible idea:
Consider using shares over reservations and limits.
● Shares balance VM resource demands rather than setting hard
thresholds.
● Less of an impact on DRS, and thus HA.
38. Big Mistake #13:
Considering Using Shares
without Considering Using Shares
• Shares are only considered during periods of contention.
• But setting Shares on Resource Pools can have
unexpected results (when you aren’t expecting them).
39. Big Mistake #13:
Considering Using Shares
without Considering Using Shares
• Shares are only considered during periods of contention.
• But setting Shares on Resource Pools can have
unexpected results (when you aren’t expecting them).
● When you assign shares to a VM,
you always specify the priority
for that virtual machine relative
to other powered-on virtual machines.
● Sibling resource pools share resources
according to their relative share values.
Simple.
Not so Simple.
40. Big Mistake #13:
Considering Using Shares
without Considering Using Shares
• Resources are divided at the Resource Pool first.
“Test” Resource Pool
1000 Shares
4 VMs
“Production” Resource Pool
4000 Shares
50 VMs
41. Big Mistake #13:
Considering Using Shares
without Considering Using Shares
• Resources are divided at the Resource Pool first.
“Test” Resource Pool
1000 Shares
4 VMs
“Production” Resource Pool
4000 Shares
50 VMs
42. Big Mistake #14:
Doing Memory Limits at All!
• Don’t assign Memory Limits. Ever.
● Let’s say you assign a VM 4G of memory.
● Then, you set a 1G memory limit on that VM.
● That VM can now never use more than 1G of physical RAM.
● All memory above 1G must come from swap or ballooning.
43. Big Mistake #14:
Doing Memory Limits at All!
• Don’t assign Memory Limits. Ever.
● Let’s say you assign a VM 4G of memory.
● Then, you set a 1G memory limit on that VM.
● That VM can now never use more than 1G of physical RAM.
● All memory above 1G must come from swap or ballooning.
• Generally best to limit memory as close to the affected
application as possible.
● Example:
Limiting SQL >
Limiting Windows >
Limiting the VM >
Limiting the Hypervisor.
44. Big Mistake #15:
Thinking You’re Smarter than DRS (‘cuz you’re not!)
• No human alive can watch every VM counter as well as
a monitor and a mathematical formula.
45. Big Mistake #16:
Not Understanding DRS’ Equations
• DRS is like a high-top table at the bar.
● Each side of that table represents a host in your cluster.
● That leg can only support the table when all sides are balanced.
● DRS’ job is to relocate VMs to ensure the table stays balanced.
46. Big Mistake #16:
Not Understanding DRS’ Equations
• DRS is like a high-top table at the bar.
● Each side of that table represents a host in your cluster.
● That leg can only support the table when all sides are balanced.
● DRS’ job is to relocate VMs to ensure the table stays balanced.
• Every five minutes a DRS interval is invoked.
● During that interval DRS analyses resource utilization counters
on every host.
● It plugs those counters into this equation:
47. Big Mistake #16:
Not Understanding DRS’ Equations
• VM entitlements
● CPU resource demand and memory working set.
● CPU and memory reservations or limits.
• Host Capacity
● Summation of CPU and memory resources, minus…
• VMKernel and Service Console overhead
• Reservations for HA Admission Control
• A small-percentage “extra” reservation
48. Big Mistake #16:
Not Understanding DRS’ Equations
• A statistical mean and standard deviation can then be
calculated.
● Mean = Average load
● Standard deviation = Average deviation from that load
49. Big Mistake #16:
Not Understanding DRS’ Equations
• A statistical mean and standard deviation can then be
calculated.
● Mean = Average load
● Standard deviation = Average deviation from that load
• This defines the Current host load standard deviation.
50. Big Mistake #16:
Not Understanding DRS’ Equations
• Your migration threshold slider value determines the
Target host load standard deviation.
51. Big Mistake #16:
Not Understanding DRS’ Equations
• DRS then runs a series of migration simulations to see
which VM moves will have the greatest impact on
balancing.
● For each simulated move, it calculates the resulting Current host
load standard deviation.
● Then, it plugs that value into this equation
52. Big Mistake #16:
Not Understanding DRS’ Equations
• The result is a priority number from 1 to 5.
● Migrations that have a greater impact on rebalancing have a
higher priority.
● Your migration threshold determines which migrations are
automatically done.
54. Big Mistake #17:
Being too Liberal.
• …with your Migration Threshold, of course.
● Migrations with lower priorities have less of an impact on
balancing our proverbial table.
● But every migration takes time, resources, and effort to
complete.
● There is a tradeoff between perfect balance and the resource
cost associated with getting to that perfect balance.
55. Big Mistake #17:
Being too Liberal.
• …with your Migration Threshold, of course.
● Migrations with lower priorities have less of an impact on
balancing our proverbial table.
● But every migration takes time, resources, and effort to
complete.
● There is a tradeoff between perfect balance and the resource
cost associated with getting to that perfect balance.
• Remember: Priority 1 recommendations are mandatory.
● These are all special cases:
• Hosts entering maintenance mode or standby mode.
• Affinity rules being violated.
• Summation of VM reservations exceed host capacity.
56. Big Mistake #18:
Combining VDI and Server Workloads
in the Same Cluster
• ESXi hosts running VDI workloads tend to experience
more load than those running server workloads.
● Greater number of concurrently running VMs.
● Workload activities are harder to predict.
● Higher frequency of VM power state changes.
• These force DRS to work harder (and more often) to
balance workloads across the cluster.
● Separating VDI workloads into their own cluster segregates this
added complexity and effort away from server activities.
57. Big Mistake #19:
Planning on Overcommit,
a.k.a. “Creating Unnecessarily Big VMs”
• Back during the “hypervisor wars” one of VMware’s big
sales points was memory overcommit.
● “ESX can overcommit memory! Hyper-V can’t!”
● So, many of us used it.
58. Big Mistake #19:
Planning on Overcommit,
a.k.a. “Creating Unnecessarily Big VMs”
• Back during the “hypervisor wars” one of VMware’s big
sales points was memory overcommit.
● “ESX can overcommit memory! Hyper-V can’t!”
● So, many of us used it.
• Overcommit creates extra work for the hypervisor.
● Ballooning, host memory swapping, page table sharing, etc.
● That work is unnecessary when memory is correctly assigned.
• Assign the right amount of memory
(and as few processors as possible) to your VMs.
● Creating “big VMs” also impacts DRS’ load balancing abilities.
● Fewer options for balancing bigger VMs.
60. Things to Remember…after the Beers…
• For the love of <your preferred deity>,
Turn on HA/DRS!
● But only if you have enough hardware!
● You’ve already paid for it.
● It is smarter than you.
61. Things to Remember…after the Beers…
• For the love of <your preferred deity>,
Turn on HA/DRS!
● But only if you have enough hardware!
● You’ve already paid for it.
● It is smarter than you.
• Understand why your VMs move around.
● Make sure that you’ve got the connected resources
they need on every host!
62. Things to Remember…after the Beers…
• For the love of <your preferred deity>,
Turn on HA/DRS!
● But only if you have enough hardware!
● You’ve already paid for it.
● It is smarter than you.
• Understand why your VMs move around.
● Make sure that you’ve got the connected resources
they need on every host!
• Save some cluster resources in reserve.
● Waste is good.
● You’ll thank me for it!
63. FILL OUT
A SURVEY
EVERY COMPLETE SURVEY
IS ENTERED INTO
DRAWING FOR A
$25 VMWARE COMPANY
STORE GIFT CERTIFICATE
64. Avoiding the 19 Biggest
HA & DRS Configuration
Mistakes - 2012 Edition
Greg Shields, Concentrated Technology
INF-VSP1232
#vmworldinf