A deep five into offloading techniques for Oracle database servers, that takes both hardware and software solution into consideration. The focus is clearly to boost the efficiency of your already paid licenses.
Building an open memory-centric computing architecture using intel optaneUniFabric
OOW 2017 presentation showcasing Fabric Attached Memory with 2 node RAC system based on two standard x86 servers running 200 GB/s with a per licensed cpu core data rate of 25 GB/s.
Converged Network Adapters have been around for a while. Now UniPlex takes things a step further giving you the very first Converged Fabric with Fabric Attached Memory, NVMe over Fabrics, SDN and PCIe Device Sharing capabilities.
Sometimes you are happy with you Desktop System, but you just need more RAM. Normally you would need to buy a Server Board ... or you try our Desktop Memory & PCIe Expansion solution and save a lot of money.
In-memory computing is a reality. So are the limits of memory capacity. Data size constantly increases, while application developers and IT staff push for in-memory efficiencies; the conclusion is inevitable: we need to be able to access more memory than the DRAM capacity that the server provides. ScaleMP’s Software Defined Memory (SDM) technology allows for more system memory to be available per server, far beyond the hardware limits, by utilizing memory from other nodes (over fabric) or from locally installed non-volatile memory (NVM) such as NAND Flash or 3D XPoint – transparently and without any changes to operating system or applications. We shall present the benefits of SDM, discuss the relevant use-cases, and share performance data.
Building an open memory-centric computing architecture using intel optaneUniFabric
OOW 2017 presentation showcasing Fabric Attached Memory with 2 node RAC system based on two standard x86 servers running 200 GB/s with a per licensed cpu core data rate of 25 GB/s.
Converged Network Adapters have been around for a while. Now UniPlex takes things a step further giving you the very first Converged Fabric with Fabric Attached Memory, NVMe over Fabrics, SDN and PCIe Device Sharing capabilities.
Sometimes you are happy with you Desktop System, but you just need more RAM. Normally you would need to buy a Server Board ... or you try our Desktop Memory & PCIe Expansion solution and save a lot of money.
In-memory computing is a reality. So are the limits of memory capacity. Data size constantly increases, while application developers and IT staff push for in-memory efficiencies; the conclusion is inevitable: we need to be able to access more memory than the DRAM capacity that the server provides. ScaleMP’s Software Defined Memory (SDM) technology allows for more system memory to be available per server, far beyond the hardware limits, by utilizing memory from other nodes (over fabric) or from locally installed non-volatile memory (NVM) such as NAND Flash or 3D XPoint – transparently and without any changes to operating system or applications. We shall present the benefits of SDM, discuss the relevant use-cases, and share performance data.
Yesterday's thinking may still believe NVMe (NVM Express) is in transition to a production ready solution. In this session, we will discuss how the evolution of NVMe is ready for production, the history and evolution of NVMe and the Linux stack to address where NVMe has progressed today to become the low latency, highly reliable database key value store mechanism that will drive the future of cloud expansion. Examples of protocol efficiencies and types of storage engines that are optimizing for NVMe will be discussed. Please join us for an exciting session where in-memory computing and persistence have evolved.
This session covers the engineering strategies and lessons learned at IBM creating industry leading in-memory data warehousing technology for use with both cloud and on-premises software. Along with rich in-memory SQL support for OLAP, data mining, and data warehousing leveraging memory optimized parallel vector processing, we’ll showcase the in-database analytics for R, spatial, and the built-in synchronization with Cloudant JSON NoSQL. We'll take a closer look at the architectural strategy for treating RAM as the new disk (and worth avoiding access to), while dramatically constraining the potential cost pressures of in-memory technology. We’ll describe how we designed for super-simplicity with load-and-go no-tuning technology for any size system, and of course… a demo. Ridiculously easy to use and freakishly fast. Not your grandmother’s IBM database.
NVMe and NVMe over fabrics promises to change the flash and networking industry. NVMe enables storage systems to tap into the full potential of flash storage and NVMe allows those systems to deliver in-server latencies. NVMe will fundamentally change storage. Are you ready? Join Storage Switzerland and Tegile for this webinar as they provide you with a path to NVMe.
Webinar: NVMe, NVMe over Fabrics and Beyond - Everything You Need to KnowStorage Switzerland
NVMe is an industry standard protocol designed specifically for memory-based storage like flash drives. It unleashes flash arrays from the chains of SCSI that hold it back. Register for our live webinar and learn what NVMe and NVMe over Fabrics are, what the advantage of NVMe is over traditional SCSI protocols and how to go beyond the NVMe basics to fully tap into the potential of memory-based storage.
A Key-Value Store for Data Acquisition SystemsIntel® Software
Get an overview of the Data Acquisition Database design. It's based on the Persistent Memory Development Kit (PMDK) and Storage Performance Development Kit (SPDK) to leverage Intel® Optane™ DC persistent memory and non-volatile memory express (NVMe) drives.
The bottleneck in flash storage is often the interface. SAS/SATA interfaces were designed specifically for hard disk drives not for flash media. For example, flash storage can support many more simultaneous I/O operations. The resolution to the problem is to use a different interface, one that is higher throughput and is more directly accessible from the CPU. Leveraging one of these interfaces and extracting optimal performance from the flash media means leaving the confines of the SCSI protocol with customized proprietary drivers. The result is complexity and slow innovation.
Bridging Big - Small, Fast - Slow with Campaign Storageinside-BigData.com
Peter Braam presented this deck at the MSST 2017 Mass Storage Conference.
"Economic considerations and technology developments are necessitating widely usable tiered storage. Untroubled by the worries of transparency and performance, Campaign Storage—invented at Los Alamos National Laboratory—offers radical revisions of old workflows and adapts to new technologies. But it also leverages widely available technologies and interfaces to offer stability from the ground up and blend in with the past. We'll discuss how a simple combination of components can support scalability, data analytics and efficient integration with memory based storage."
Peter Braam is a scientist and entrepreneur focused on large scale computing problems. After obtaining a PhD in mathematics under Michael Atiyah, he was an academic at several universities including Oxford, CMU and Cambridge. One of his startup companies developed the Lustre file system which is widely used. Most other products he designed were sold to major corporations. From 2013, Peter has been assisting computing design for the SKA telescope as a consultant. Currently Peter is doing research in storage and also architecting a product for Campaign Storage, LLC.
Watch the video: http://wp.me/p3RLHQ-gNC
Learn more: http://campaignstorage.com/
and
http://storageconference.us/
Sign up for our insideHPC Newsletter: http://insidehpc.com/newsletter
IBM POWER8 processor is the fastest available on the market, redefining Open Source performance. With this amazing processor, IBM and members of the OpenPower Foundation design innovative and cost-effective systems, delivering the infrastructure of choice for the most demanding workloads, in terms of throughput, scalability and reliability.
In this talk in english, Thibaud Besson will browse the key characteristics of Power Systems, why they are the most relevant for today's challenges, both from a technical and economical standpoint. Finally, we will review the possibilities you have to get your hands on one of these outstanding plateforms for your Open Source applications.
Yesterday's thinking may still believe NVMe (NVM Express) is in transition to a production ready solution. In this session, we will discuss how the evolution of NVMe is ready for production, the history and evolution of NVMe and the Linux stack to address where NVMe has progressed today to become the low latency, highly reliable database key value store mechanism that will drive the future of cloud expansion. Examples of protocol efficiencies and types of storage engines that are optimizing for NVMe will be discussed. Please join us for an exciting session where in-memory computing and persistence have evolved.
This session covers the engineering strategies and lessons learned at IBM creating industry leading in-memory data warehousing technology for use with both cloud and on-premises software. Along with rich in-memory SQL support for OLAP, data mining, and data warehousing leveraging memory optimized parallel vector processing, we’ll showcase the in-database analytics for R, spatial, and the built-in synchronization with Cloudant JSON NoSQL. We'll take a closer look at the architectural strategy for treating RAM as the new disk (and worth avoiding access to), while dramatically constraining the potential cost pressures of in-memory technology. We’ll describe how we designed for super-simplicity with load-and-go no-tuning technology for any size system, and of course… a demo. Ridiculously easy to use and freakishly fast. Not your grandmother’s IBM database.
NVMe and NVMe over fabrics promises to change the flash and networking industry. NVMe enables storage systems to tap into the full potential of flash storage and NVMe allows those systems to deliver in-server latencies. NVMe will fundamentally change storage. Are you ready? Join Storage Switzerland and Tegile for this webinar as they provide you with a path to NVMe.
Webinar: NVMe, NVMe over Fabrics and Beyond - Everything You Need to KnowStorage Switzerland
NVMe is an industry standard protocol designed specifically for memory-based storage like flash drives. It unleashes flash arrays from the chains of SCSI that hold it back. Register for our live webinar and learn what NVMe and NVMe over Fabrics are, what the advantage of NVMe is over traditional SCSI protocols and how to go beyond the NVMe basics to fully tap into the potential of memory-based storage.
A Key-Value Store for Data Acquisition SystemsIntel® Software
Get an overview of the Data Acquisition Database design. It's based on the Persistent Memory Development Kit (PMDK) and Storage Performance Development Kit (SPDK) to leverage Intel® Optane™ DC persistent memory and non-volatile memory express (NVMe) drives.
The bottleneck in flash storage is often the interface. SAS/SATA interfaces were designed specifically for hard disk drives not for flash media. For example, flash storage can support many more simultaneous I/O operations. The resolution to the problem is to use a different interface, one that is higher throughput and is more directly accessible from the CPU. Leveraging one of these interfaces and extracting optimal performance from the flash media means leaving the confines of the SCSI protocol with customized proprietary drivers. The result is complexity and slow innovation.
Bridging Big - Small, Fast - Slow with Campaign Storageinside-BigData.com
Peter Braam presented this deck at the MSST 2017 Mass Storage Conference.
"Economic considerations and technology developments are necessitating widely usable tiered storage. Untroubled by the worries of transparency and performance, Campaign Storage—invented at Los Alamos National Laboratory—offers radical revisions of old workflows and adapts to new technologies. But it also leverages widely available technologies and interfaces to offer stability from the ground up and blend in with the past. We'll discuss how a simple combination of components can support scalability, data analytics and efficient integration with memory based storage."
Peter Braam is a scientist and entrepreneur focused on large scale computing problems. After obtaining a PhD in mathematics under Michael Atiyah, he was an academic at several universities including Oxford, CMU and Cambridge. One of his startup companies developed the Lustre file system which is widely used. Most other products he designed were sold to major corporations. From 2013, Peter has been assisting computing design for the SKA telescope as a consultant. Currently Peter is doing research in storage and also architecting a product for Campaign Storage, LLC.
Watch the video: http://wp.me/p3RLHQ-gNC
Learn more: http://campaignstorage.com/
and
http://storageconference.us/
Sign up for our insideHPC Newsletter: http://insidehpc.com/newsletter
IBM POWER8 processor is the fastest available on the market, redefining Open Source performance. With this amazing processor, IBM and members of the OpenPower Foundation design innovative and cost-effective systems, delivering the infrastructure of choice for the most demanding workloads, in terms of throughput, scalability and reliability.
In this talk in english, Thibaud Besson will browse the key characteristics of Power Systems, why they are the most relevant for today's challenges, both from a technical and economical standpoint. Finally, we will review the possibilities you have to get your hands on one of these outstanding plateforms for your Open Source applications.
Realizing Exabyte-scale PM Centric Architectures and Memory Fabricsinside-BigData.com
In this video from the SNIA Persistent Memory Summit, Zvonimir Bandic from Western Digital presents: Realizing the Next Generation of Exabyte-scale PM Centric Architectures and Memory Fabrics.
In the last five years, the increasing volume, velocity and variety of data generated and consumed by Big Data and Fast Data applications has driven an aggressive pursuit for the next generation of emerging non-volatile memories, particularly in the area of persistent memory. At component level, this memory must be byte-addressable and non-volatile, deliver latency comparable to DRAM, but have density and cost that falls somewhere between DRAM and NAND flash.
Much has been debated about would it take to scale a system to exabyte main memory with the right levels of latencies to address the world’s growing and diverse data needs. This presentation will explore legacy distributed system architectures based on traditional CPU and peripheral attachment of persistent memory, scaled out through the use of RDMA networking. It will discuss the present boundaries of memory and compute technologies, and the many considerations for developing persistent memory, including performance, power, latency requirements and cost merits of parallel and serial attachment points for memories, and show the experimentally measured latency of RDMA access to persistent memory devices.
This presentation will also consider a theoretical question of what would it take to scale a system to exabyte main memory from the perspective of networking fabric required to access such large amounts of main memory at useful latencies. It will explore the “exabyte challenge” from the hardware architecture perspective and, given the present boundaries of memory and compute technologies, quantitatively evaluate latency requirements for memory and memory fabric switch devices. In addition, it will address the ramifications of the large memory footprint of persistent memory for emerging data-intensive workloads, such as high performance data analytics, autonomous vehicles, social networking value extraction, and many traditional memory bound workloads. Finally, it will outline a vision for a prototyping platform for accelerating innovation in networking protocols that will enable experimental evaluation of novel memory fabrics at scale.
Watch the video: https://wp.me/p3RLHQ-i2k
Sign up for our insideHPC Newsletter: http://insidehpc.com/newsletter
Технологии работы с дисковыми хранилищами и файловыми системами Windows Serve...Виталий Стародубцев
##Что такое Storage Replica
##Архитектура и сценарии
##Синхронная и асинхронная репликация
##Междисковая, межсерверная, внутрикластерная и межкластерная репликация
##Дизайн и проектирование Storage Replica
##Нововведения в Windows Server 2016 TP5
##Графический интерфейс управления, и другие возможности - демонстрация и планы развития
##Интеграция Storage Replica с Storage Spaces Direct
Optimized HPC/AI cloud with OpenStack acceleration service and composable har...Shuquan Huang
Today data scientist is turning to cloud for AI and HPC workloads. However, AI/HPC applications require high computational throughput where generic cloud resources would not suffice. There is a strong demand for OpenStack to support hardware accelerated devices in a dynamic model.
In this session, we will introduce OpenStack Acceleration Service – Cyborg, which provides a management framework for accelerator devices (e.g. FPGA, GPU, NVMe SSD). We will also discuss Rack Scale Design (RSD) technology and explain how physical hardware resources can be dynamically aggregated to meet the AI/HPC requirements. The ability to “compose on the fly” with workload-optimized hardware and accelerator devices through an API allow data center managers to manage these resources in an efficient automated manner.
We will also introduce an enhanced telemetry solution with Gnnochi, bandwidth discovery and smart scheduling, by leveraging RSD technology, for efficient workloads management in HPC/AI cloud.
Slide chia sẻ công nghệ về caching, thông qua slide này bạn sẽ trả lời được những câu hỏi như:
- Caching là gì
- Làm sao sử dụng cũng như xây dựng hệ thống caching
- Tại sao cache giúp tăng tốc ứng dụng lên vài chục, vài trăm lần
- Các hệ thống lớn của Facebook, Twitter, ... đang sử dụng cache thế nào
- ...
Slide chia sẻ về công nghệ về caching, thông qua slide này bạn sẽ trả lời được những câu hỏi như:
- Caching là gì
- Làm sao sử dụng cũng như xây dựng hệ thống caching
- Tại sao cache giúp tăng tốc ứng dụng lên vài chục, vài trăm lần
- Các hệ thống lớn của Facebook, Twitter, ... đang sử dụng cache thế nào
- ...
Running Apache Spark on a High-Performance Cluster Using RDMA and NVMe Flash ...Databricks
Effectively leveraging fast networking and storage hardware (e.g., RDMA, NVMe, etc.) in Apache Spark remains challenging. Current ways to integrate the hardware at the operating system level fall short, as the hardware performance advantages are shadowed by higher layer software overheads. This session will show how to integrate RDMA and NVMe hardware in Spark in a way that allows applications to bypass both the operating system and the Java virtual machine during I/O operations. With such an approach, the hardware performance advantages become visible at the application level, and eventually translate into workload runtime improvements. Stuedi will demonstrate how to run various Spark workloads (e.g, SQL, Graph, etc.) effectively on 100Gbit/s networks and NVMe flash.
DPDK Summit 2015 - Aspera - Charles ShiflettJim St. Leger
DPDK Summit 2015 in San Francisco.
Presentation by Charles Shiflett, Aspera.
For additional details and the video recording please visit www.dpdksummit.com.
Crimson: Ceph for the Age of NVMe and Persistent MemoryScyllaDB
Ceph is a mature open source software-defined storage solution that was created over a decade ago.
During that time new faster storage technologies have emerged including NVMe and Persistent memory.
The crimson project aim is to create a better Ceph OSD that is more well suited to those faster devices. The crimson OSD is built on the Seastar C++ framework and can leverage these devices by minimizing latency, cpu overhead, and cross-core communication. This talk will discuss the project design, our current status, and our future plans.
Enhanced Enterprise Intelligence with your personal AI Data Copilot.pdfGetInData
Recently we have observed the rise of open-source Large Language Models (LLMs) that are community-driven or developed by the AI market leaders, such as Meta (Llama3), Databricks (DBRX) and Snowflake (Arctic). On the other hand, there is a growth in interest in specialized, carefully fine-tuned yet relatively small models that can efficiently assist programmers in day-to-day tasks. Finally, Retrieval-Augmented Generation (RAG) architectures have gained a lot of traction as the preferred approach for LLMs context and prompt augmentation for building conversational SQL data copilots, code copilots and chatbots.
In this presentation, we will show how we built upon these three concepts a robust Data Copilot that can help to democratize access to company data assets and boost performance of everyone working with data platforms.
Why do we need yet another (open-source ) Copilot?
How can we build one?
Architecture and evaluation
Adjusting primitives for graph : SHORT REPORT / NOTESSubhajit Sahu
Graph algorithms, like PageRank Compressed Sparse Row (CSR) is an adjacency-list based graph representation that is
Multiply with different modes (map)
1. Performance of sequential execution based vs OpenMP based vector multiply.
2. Comparing various launch configs for CUDA based vector multiply.
Sum with different storage types (reduce)
1. Performance of vector element sum using float vs bfloat16 as the storage type.
Sum with different modes (reduce)
1. Performance of sequential execution based vs OpenMP based vector element sum.
2. Performance of memcpy vs in-place based CUDA based vector element sum.
3. Comparing various launch configs for CUDA based vector element sum (memcpy).
4. Comparing various launch configs for CUDA based vector element sum (in-place).
Sum with in-place strategies of CUDA mode (reduce)
1. Comparing various launch configs for CUDA based vector element sum (in-place).
Analysis insight about a Flyball dog competition team's performanceroli9797
Insight of my analysis about a Flyball dog competition team's last year performance. Find more: https://github.com/rolandnagy-ds/flyball_race_analysis/tree/main
Adjusting OpenMP PageRank : SHORT REPORT / NOTESSubhajit Sahu
For massive graphs that fit in RAM, but not in GPU memory, it is possible to take
advantage of a shared memory system with multiple CPUs, each with multiple cores, to
accelerate pagerank computation. If the NUMA architecture of the system is properly taken
into account with good vertex partitioning, the speedup can be significant. To take steps in
this direction, experiments are conducted to implement pagerank in OpenMP using two
different approaches, uniform and hybrid. The uniform approach runs all primitives required
for pagerank in OpenMP mode (with multiple threads). On the other hand, the hybrid
approach runs certain primitives in sequential mode (i.e., sumAt, multiply).
4. Definition of offloading (DB view)
In general:
«Everything, that saves resources on
the database server»
5. Definition of offloading (DB view)
Examples of offloading implementations
NIC (TCP/IP Offload, iSCSI Offload, Infiniband RDMA, NVMe)
Storage Adatapets (RAID Calculation, SCSI)
Math Co-Processors
FPGAs
DMA-Engines
Distributed Computing (e.g. using MPI)
Remote DB Engine (Hadoop Connector, Gluent)
6. Definition of offloading (DB view)
How is it done the Exadata?
Offloading via DMA-Engine of the Infiniband HCA
Enables Remote-DMA (RDMA) Operations (DB to Cell)
The storage cell can be acessed at near zero cpu cost
Latency of a DMA operation is higher than PIO via CPU therefore good for large
amounts of data e.g. DWH, but worse for OLTP
The task can be distributed
Order e.g. to execute a sub-query on a node via MPI-call and to transmit the start
or end memory address to the requester (DB server)
The DB server now only needs to merge the partial results.
The DB server is in this sense more acting as a client
7. Offloading techniques we can use
The following devices have a DMA engine:
RDMA-enabled network adapters and Infiniband cards
Intel IOATDMA chip on Xeon boards (for NVMe SSDs
PCIe switch cards
PLX-based NVMe controllers
Or the PCIe chip in your Intel Xeon computer ;-)
Lowest latency
8. Offloading techniques we can use
The following protocols have (R) DMA support:
iSCSI over RMDA
NFS over RDMA
NVMe over Fabrics (RDMA-based) or RDMA Block Device
Needs the least CPU
Good starting point
9. Offloading techniques we can use
Comparison (Native PCIe fabric vs. NVMe over Fabrics)
Native PCIe fabric has significantly less latency
Setup with PCIe-JBOF is less complex than NVMe over Fabrics
Throughput is identical
10. Offloading techniques we can use
That PCIe is quite cool… What other tricks can it do?
DMA-Engine like Infiniband
Connect multiple PCIe root complexes via Non-Transparent Bridge
Network protocol IPoPCIe analogous to IPoIB, but performs way better
Device Sharing via I / O Virtualization (SR-IOV, MR-IOV)
11. Offloading techniques we can use
How do we get the system really fast?
Answer: Memory!
The only question is:
Which memory?
Where is it located?
How is it structured?
12. Demo-Time ☺
Demo 1: Device Sharing
Description
Host 1 has a SR-IOV capable NIC
Host 1 initializes a Virtual Function
Through Non-Transparent Bridge
(NTB) Host 2 can access that
function by loading the device driver
for the NIC
https://www.youtube.com/watch?v=GPh0Ms3dfPo
13. Demo-Time ☺
Demo 1: Device Sharing
Expected behaviour
Works as designed ☺
Depending on the approach PCIe switch chip, there is device driver dependencies
14. Demo-Time ☺
Demo 2: DMA-Transfer
Description
Host 1 and Host2 are fitted with a
PCIe Switch based host card and
connected back to back
PLXSDK comes with a Sample
Program supporting PIO and DMA
transfer
We measure the overall throughput
and cpu load
https://www.youtube.com/watch?v=LNPBr3WvuNg
15. Demo-Time ☺
Demo 2: DMA-Transfer
Expected behaviour
Large data transfer benefits from DMA (DWH) ☺
Small, time critical transfers have less latency with PIO (OLTP)
You’ll need both modes
16. Demo-Time ☺
Demo 3: Fabric Attached Memory (PCIe) and Oracle RAC
Description
Database and Memory hosts are fitted
with a PCIe Switch based host card and
connected to a central PCIe Switch
Memory hosts’s physical DRAM is
expanded with OptaneGrid 3DXpoint
into an SDM Pool (mirrored via PCIe
NTB)
Database Servers expose a tiered
PMEM Device using local DRAM
(mirrored via PCIe NTB) and the remote
SDM Pool accessed over PCIe NTB)
ASM High Redudancy on top of PMEM
Devices with preferred mirror read and
device mapper path swapping
db0 db1 db2
mem0 mem1 mem2
SDM
DRAM
Optane
GRID
SDM
DRAM
Optane
GRID
SDM
DRAM
Optane
GRID
ASM
PMEM
DRAM
Expansion
PMEM
DRAM
Expansion
PMEM
DRAM
Expansion
PCIe Switch
RAC
NTB
Domain
17. Demo-Time ☺
Demo 3: Fabric Attached Memory (PCIe) and Oracle RAC
16 GB/s throughput per licensable core (4cores, 8 threads per db node)
85 % of native aggregated memory controller performance
18. Findings
Generic offloading is possible per se, but different than expected :
Fabric Attached Memory
Yes, the DB is running in memory (mirrored)
Question is:
In which server’s memory (local or remote)?
How do we acccess it (local memory extension or DMA call)?
How is it constructed (DRAM or Software Defined Memory)?
Using the right PCIe-Switch and storage module combination you
get it to work
Any PCIe-capable host can use Fabric Attached Memory per se
An OpenMCCA-compatible PCIe switch (PLX 9700) and high-performance M.2 SSDs
such as Optane Memory or fast NVMe modules are required