SlideShare a Scribd company logo
1 of 29
Download to read offline
Optimistic Offer -
What does it mean to Mesos Frameworks?
Yong Feng
MesosCon North America 2016, 6/01/2016
1
yongfeng@ca.ibm.com
Overview
• The Pessimistic Offer Programming Model
• Implementing frameworks to use Pessimistic Offer
• Observed limitations and proposed mitigations
• The Optimistic Offer Programming Model
• Benefits
• Design Plan
• Development status
• Optimistic vs Pessimistic Offer
• Discussion and Additional Resources
2
The Pessimistic Offer Programming Model
• Offer: allocate available resources in a host to a single framework. Allocated
resources are locked (cached) for a period of time.
• Resources are allocated or partitioned among frameworks without any knowledge of a
framework’s requirements
• Resources in an offer is either consumed by launching task, or freed by rejecting or
rescinding of an offer
• Frameworks cannot consume or reject parts of an offer while still keeping the left over
resources of that offer
• Resources are only available for re-allocation after the offer is rejected/rescinded or the
task/executor is finished
• Inverse Offer: deallocate (or reclaim) resources from a framework
3
Free Allocated Used
Framework launch task
Reject or rescind
Offer to framework
Task or executor finished
Framework Implementation: Cache Offers
4
• What if a framework has no suitable demand when an offer is received?
• They cache it !
• Caching resources can lead to:
• Better performance for future demand
• Better scheduling decision with more candidate hosts, for example affinity …
• Stockpiled resources for future tasks that have higher demands
• Still want to be a good citizen?
• Allow for TTL (expiration) of cached offers
• Frameworks that support TTL of cached offers:
• Swarm
• Kubernetes
Framework Implementation: Revive Offers
5
• What if there is no demand to consume the offers in a framework?
• Reject it ! (or send suppress request)
• Rejecting or suppressing unwanted resources allow for:
• Better resource utilization by allocating resources to frameworks that have a demand
• What if there is a demand in a framework, but no offer?
• Revive the offer
• Send a request to Mesos to reallocate the offer
• Frameworks that support Suppress/Revive offer:
• Marathon
Pessimistic Offer: Limitations
Inefficient resource allocation Mitigation
Cached resource are not released to other
frameworks in time
Over a period of time, each offer only includes
small pieces of each resource in a host. To fix
this, offer TTL must be enforced by Mesos and
Mesos frameworks.
PO – Pessimistic Offer
OO – Optimistic Offer
Average time to receive offers (ms) for multiple client frameworks (Pessimistic vs Optimistic Offer)
6
Pessimistic Offer: Limitations (Cont’d)
Poor scheduling quality Mitigation
Offered resource may not be the most suitable
resource
Frameworks should delay the scheduling decision
until they get suitable resources
PO – Pessimistic Offer
OO – Optimistic Offer
Job duration time (ms) for multiple client frameworks (Pessimistic vs Optimistic Offers)
7
Pessimistic Offer: Limitations (Cont’d)
Low QoS guarantee Mitigation
Long tasks prevent resource re-allocation.
Hard to preempt resource among frameworks.
Implement Quotas, reservations, over-
subscription …
https://www.youtube.com/watch?v=jC8lhGQN2Sc 8
Can We Squeeze More Out of Pessimistic Offers?
9
• Adjust the size of an offer (Mesos-3765)
• Smaller sized offers improves resource utilization, and fairness
• Oversubscription for reservation (Mesos-4967)
• Lend reserved resource to other framework if it is not being used. This improves
resource utilization without impacting QoS
• Reuse the recovered resource without delay (Mesos-3078 or Mesos-4811)
• Accelerate scheduling so that resources can be used once it becomes available
• More hints in filter when rejecting offer
• Avoid ping-pong of resources between Mesos and framework to improve performance
and avoid starvation
The Optimistic Offer Model
Offer: displays all available resources to multiple frameworks.
Benefits of the Optimistic Offer Model
• Better resource utilization
• Available resources are always visible to all frameworks
• Improved scheduling performance
• Resources recovered from finished tasks are visible to all frameworks immediately
• Enhanced quality of scheduling decision
• Increased availability of resources help frameworks to make better scheduling
decisions
• QoS guarantee
• QoS of workload is enforced by pre-emption
11
The Optimistic Offer - Design Plan
• Track both offered resources and consumed resources
• Offered resources are not allocated resources. Offered resources are viewed as available resources to all
framework
• Consumed resources are equivalent to allocated resource in the Pessimistic Offer model, however they
might still be viewed as available resources to some frameworks under special conditions.
• Offered resources are based on resource plan or usage instead of allocation
• Available resources are visible to frameworks via Offer
• Available resources are decided by resource plan and resource usage
• Refresh resource and quota availability in a timely manner
• Update resource availability and status to framework by either rejecting offer from framework or rescinding
offer from master
• Update resource plan availability with quota or restrictions to each framework
• Pre-empt tasks to enforce the QoS
• Pre-empt task from lower priority framework or overused framework to enforce QoS
• Export the resource status to framework to achieve “smart” pre-emption
12
The Optimistic Offer - Design Plan (Cont’d)
• Programming model:
• Allow frameworks to use offers without
the need for rejection of the offer from
a prior framework
• Refresh the status of a resource by
handling rescind messages or rejected
offers
• Use the non-revocable resource first
and then revocable resource.
• Handle the inverse offer smartly for
pre-emption
13
The Optimistic Offer - Development Status
• Mesos-1607 Phase 1 (renamed as “Oversubscription for reservation” in
Mesos-4967)
• Target: ~4Q 2016
• Status: Reviewable; ~30 patches are ready for review, try it after applying
the patches
14
Total Resources
Reserved
Allocated
Used
Revocable
The Optimistic Offer - Development Status (Cont’d)
15
• Mesos-1607 Prototype
• Target: IBM Mesos Connector
• Design: adjust allocator API to track consumed resources and resolve resource races
• Status: In Bluemix Container Cloud Service, and planned in Mesos roadmap.
Optimistic Offer vs Pessimistic Offer Model
16
Idle frameworks Idle frameworks with smaller tasks
PO – Pessimistic Offer
OO – Optimistic Offer
(Number of task per ms)
Optimistic Offer vs Pessimistic Offer Model
17
PO – Pessimistic Offer
OO – Optimistic Offer
(Number of task per ms)
Idle frameworks Idle frameworks with smaller tasks and small TTL
Optimistic Offer vs Pessimistic Offer Model
18
PO – Pessimistic Offer
OO – Optimistic Offer
(Number of task per ms)
Idle frameworks Idle frameworks when placement constraint is added
Optimistic Offer vs Pessimistic Offer Model
19
PO – Pessimistic Offer
OO – Optimistic Offer
(Job duration per ms for multiple clients)
Idle frameworks Idle frameworks with smaller tasks
Optimistic Offer vs Pessimistic Offer Model
20
PO – Pessimistic Offer
OO – Optimistic Offer
(Job duration per ms for multiple clients)
Idle frameworks Idle frameworks with smaller tasks and smaller TTL
Optimistic Offer vs Pessimistic Offer Model
21
PO – Pessimistic Offer
OO – Optimistic Offer
(slave utilization)
Idle frameworks Idle frameworks with smaller tasks
Optimistic Offer vs Pessimistic Offer Model
22
PO – Pessimistic Offer
OO – Optimistic Offer
(slave utilization)
Idle frameworks Idle frameworks with smaller tasks and smaller TTL
Evaluation – Conflicts
• 4 swarm frameworks
• 10 nodes, 1000 containers (tasks)
• Task request rejected after 10 placement retries
(a) System load = 0.8 and
varied request resource size
(b) System load = 0.4,0.8,0.99
and random request resource size
0
5
10
15
20
25
small medium large varied
Percentageoftasks
Task resource request size
Conflicted tasks (%)
Rejected tasks (%)
0
2
4
6
8
10
12
14
16
40% 80% 99%
Percentageoftasks
System Load
Conflicted tasks (%)
Rejected tasks (%)
Optimistic Offer in IBM Bluemix Container Cloud
24
Mesos Master
Swarm
Scheduler
Docker
CLI/API
framework
Kubernetes
Scheduler
Mesos
Agents
Offers
framework
Offers
Tasks to Mesos
Kubernetes
CLI/API
Mesos
Agents
Mesos
Agents
Network
Agents
Network
Agents
Tasks to Mesos
Open Questions
• How to export more policy status and configuration to
frameworks so that frameworks can make better decisions?
• How to reduce scheduling overheads, such as resource
conflicts?
• …
25
IBM Booth
26
Welcome to S5
• Spark benchmark with IBM Session Scheduler
• Bluemix Container Service
• GPU in Power
• Mesos Connector
Mesos Community Activities
27
• Active development with Mesos community
– 11 IBM Developers.
• 100+ JIRAs delivered or in progress
• Leading or participating in several work
streams: POWER Support, Optimistic
Offers, Container Support, GPU Support,
Swarm and Kubernetes integration
• Relationship with Mesosphere – weekly
calls, on-site developer presence
• Attendance at MesosCon 2016 with
sponsorship and booth
• Aligning with IBM Container Service to
leverage common OSS technologies
• Technical Preview of Mesos with IBM
Value-Add (ASC) on Docker Hub – Both
x86 and POWER images
IBM Committed Mesos Patches
28
For the period June 2015 to May 2016
Thank you

More Related Content

What's hot

Resource Management in Impala - StampedeCon 2016
Resource Management in Impala - StampedeCon 2016Resource Management in Impala - StampedeCon 2016
Resource Management in Impala - StampedeCon 2016StampedeCon
 
Reservations Based Scheduling: if you’re late don’t blame us!
Reservations Based Scheduling: if you’re late don’t blame us!  Reservations Based Scheduling: if you’re late don’t blame us!
Reservations Based Scheduling: if you’re late don’t blame us! DataWorks Summit
 
Admission Control in Impala
Admission Control in ImpalaAdmission Control in Impala
Admission Control in ImpalaCloudera, Inc.
 
Beginners Guide to High Availability for Postgres
Beginners Guide to High Availability for PostgresBeginners Guide to High Availability for Postgres
Beginners Guide to High Availability for PostgresEDB
 
Logical Architecture for Protection
Logical Architecture for ProtectionLogical Architecture for Protection
Logical Architecture for ProtectionSunita Shrivastava
 
Writing Yarn Applications Hadoop Summit 2012
Writing Yarn Applications Hadoop Summit 2012Writing Yarn Applications Hadoop Summit 2012
Writing Yarn Applications Hadoop Summit 2012Hortonworks
 
Hyperion Planning System 9 Installation Guide
Hyperion Planning   System 9 Installation GuideHyperion Planning   System 9 Installation Guide
Hyperion Planning System 9 Installation GuideShehzad Kazmi
 
How did you know this ad would be relevant for me?
How did you know this ad would be relevant for me?How did you know this ad would be relevant for me?
How did you know this ad would be relevant for me?DataWorks Summit
 
YARN - Next Generation Compute Platform fo Hadoop
YARN - Next Generation Compute Platform fo HadoopYARN - Next Generation Compute Platform fo Hadoop
YARN - Next Generation Compute Platform fo HadoopHortonworks
 
How AOL Advertising Uses NoSQL to Make Millions of Smart Targeting Decisions ...
How AOL Advertising Uses NoSQL to Make Millions of Smart Targeting Decisions ...How AOL Advertising Uses NoSQL to Make Millions of Smart Targeting Decisions ...
How AOL Advertising Uses NoSQL to Make Millions of Smart Targeting Decisions ...DATAVERSITY
 
How to scale recommendation system with HBase
How to scale recommendation system with HBaseHow to scale recommendation system with HBase
How to scale recommendation system with HBaseRafael Arana
 
Next Generation Hadoop: High Availability for YARN
Next Generation Hadoop: High Availability for YARN Next Generation Hadoop: High Availability for YARN
Next Generation Hadoop: High Availability for YARN Arinto Murdopo
 

What's hot (13)

Resource Management in Impala - StampedeCon 2016
Resource Management in Impala - StampedeCon 2016Resource Management in Impala - StampedeCon 2016
Resource Management in Impala - StampedeCon 2016
 
Reservations Based Scheduling: if you’re late don’t blame us!
Reservations Based Scheduling: if you’re late don’t blame us!  Reservations Based Scheduling: if you’re late don’t blame us!
Reservations Based Scheduling: if you’re late don’t blame us!
 
Admission Control in Impala
Admission Control in ImpalaAdmission Control in Impala
Admission Control in Impala
 
Beginners Guide to High Availability for Postgres
Beginners Guide to High Availability for PostgresBeginners Guide to High Availability for Postgres
Beginners Guide to High Availability for Postgres
 
Logical Architecture for Protection
Logical Architecture for ProtectionLogical Architecture for Protection
Logical Architecture for Protection
 
Virtualized Hadoop
Virtualized HadoopVirtualized Hadoop
Virtualized Hadoop
 
Writing Yarn Applications Hadoop Summit 2012
Writing Yarn Applications Hadoop Summit 2012Writing Yarn Applications Hadoop Summit 2012
Writing Yarn Applications Hadoop Summit 2012
 
Hyperion Planning System 9 Installation Guide
Hyperion Planning   System 9 Installation GuideHyperion Planning   System 9 Installation Guide
Hyperion Planning System 9 Installation Guide
 
How did you know this ad would be relevant for me?
How did you know this ad would be relevant for me?How did you know this ad would be relevant for me?
How did you know this ad would be relevant for me?
 
YARN - Next Generation Compute Platform fo Hadoop
YARN - Next Generation Compute Platform fo HadoopYARN - Next Generation Compute Platform fo Hadoop
YARN - Next Generation Compute Platform fo Hadoop
 
How AOL Advertising Uses NoSQL to Make Millions of Smart Targeting Decisions ...
How AOL Advertising Uses NoSQL to Make Millions of Smart Targeting Decisions ...How AOL Advertising Uses NoSQL to Make Millions of Smart Targeting Decisions ...
How AOL Advertising Uses NoSQL to Make Millions of Smart Targeting Decisions ...
 
How to scale recommendation system with HBase
How to scale recommendation system with HBaseHow to scale recommendation system with HBase
How to scale recommendation system with HBase
 
Next Generation Hadoop: High Availability for YARN
Next Generation Hadoop: High Availability for YARN Next Generation Hadoop: High Availability for YARN
Next Generation Hadoop: High Availability for YARN
 

Viewers also liked

E.S.T. No. 1 MIGUEL LERDO DE TEJADA, ESPACIO MAKER 3 A
E.S.T. No. 1 MIGUEL LERDO DE TEJADA, ESPACIO MAKER 3 AE.S.T. No. 1 MIGUEL LERDO DE TEJADA, ESPACIO MAKER 3 A
E.S.T. No. 1 MIGUEL LERDO DE TEJADA, ESPACIO MAKER 3 AVJ Medina Ortiz
 
PLS 2016: ILP Technical Committee latest expert guidance & documents – PLG08
PLS 2016: ILP Technical Committee latest expert guidance & documents – PLG08PLS 2016: ILP Technical Committee latest expert guidance & documents – PLG08
PLS 2016: ILP Technical Committee latest expert guidance & documents – PLG08Institution of Lighting Professionals
 
The Three Things
The Three ThingsThe Three Things
The Three ThingsAgileDenver
 
Track 3 session 8 - st dev con 2016 - music and voice over ble
Track 3   session 8 - st dev con 2016 - music and voice over bleTrack 3   session 8 - st dev con 2016 - music and voice over ble
Track 3 session 8 - st dev con 2016 - music and voice over bleST_World
 
Real World Use Case with Cassandra (Eddie Satterly, DataNexus) | C* Summit 2016
Real World Use Case with Cassandra (Eddie Satterly, DataNexus) | C* Summit 2016Real World Use Case with Cassandra (Eddie Satterly, DataNexus) | C* Summit 2016
Real World Use Case with Cassandra (Eddie Satterly, DataNexus) | C* Summit 2016DataStax
 
Kurumsal Ağlarda Saldırı Tespiti Amaçlı Honeypot Sistemlerin Efektif Kullanımı
Kurumsal Ağlarda Saldırı Tespiti Amaçlı Honeypot Sistemlerin Efektif KullanımıKurumsal Ağlarda Saldırı Tespiti Amaçlı Honeypot Sistemlerin Efektif Kullanımı
Kurumsal Ağlarda Saldırı Tespiti Amaçlı Honeypot Sistemlerin Efektif KullanımıBGA Cyber Security
 
Gerçek Dünyadan Siber Saldırı Örnekleri
Gerçek Dünyadan Siber Saldırı Örnekleri Gerçek Dünyadan Siber Saldırı Örnekleri
Gerçek Dünyadan Siber Saldırı Örnekleri BGA Cyber Security
 
Czy mogą się udawać projekty publiczne?
Czy mogą się udawać projekty publiczne?Czy mogą się udawać projekty publiczne?
Czy mogą się udawać projekty publiczne?DAT Computer Concepts
 
CES 2016 – GfK smart home presentation
CES 2016 – GfK smart home presentationCES 2016 – GfK smart home presentation
CES 2016 – GfK smart home presentationGfK
 
Mobile Application Penetration Testing
Mobile Application Penetration TestingMobile Application Penetration Testing
Mobile Application Penetration TestingBGA Cyber Security
 
Track 4 session 5 - st dev con 2016 - simplifying the setup and use of iot ...
Track 4   session 5 - st dev con 2016 - simplifying the setup and use of iot ...Track 4   session 5 - st dev con 2016 - simplifying the setup and use of iot ...
Track 4 session 5 - st dev con 2016 - simplifying the setup and use of iot ...ST_World
 

Viewers also liked (15)

E.S.T. No. 1 MIGUEL LERDO DE TEJADA, ESPACIO MAKER 3 A
E.S.T. No. 1 MIGUEL LERDO DE TEJADA, ESPACIO MAKER 3 AE.S.T. No. 1 MIGUEL LERDO DE TEJADA, ESPACIO MAKER 3 A
E.S.T. No. 1 MIGUEL LERDO DE TEJADA, ESPACIO MAKER 3 A
 
Dares apprentissage en 2015
Dares apprentissage en 2015Dares apprentissage en 2015
Dares apprentissage en 2015
 
PLS 2016: ILP Technical Committee latest expert guidance & documents – PLG08
PLS 2016: ILP Technical Committee latest expert guidance & documents – PLG08PLS 2016: ILP Technical Committee latest expert guidance & documents – PLG08
PLS 2016: ILP Technical Committee latest expert guidance & documents – PLG08
 
Agile in Offshore development
Agile in Offshore developmentAgile in Offshore development
Agile in Offshore development
 
The Three Things
The Three ThingsThe Three Things
The Three Things
 
Track 3 session 8 - st dev con 2016 - music and voice over ble
Track 3   session 8 - st dev con 2016 - music and voice over bleTrack 3   session 8 - st dev con 2016 - music and voice over ble
Track 3 session 8 - st dev con 2016 - music and voice over ble
 
Real World Use Case with Cassandra (Eddie Satterly, DataNexus) | C* Summit 2016
Real World Use Case with Cassandra (Eddie Satterly, DataNexus) | C* Summit 2016Real World Use Case with Cassandra (Eddie Satterly, DataNexus) | C* Summit 2016
Real World Use Case with Cassandra (Eddie Satterly, DataNexus) | C* Summit 2016
 
Kurumsal Ağlarda Saldırı Tespiti Amaçlı Honeypot Sistemlerin Efektif Kullanımı
Kurumsal Ağlarda Saldırı Tespiti Amaçlı Honeypot Sistemlerin Efektif KullanımıKurumsal Ağlarda Saldırı Tespiti Amaçlı Honeypot Sistemlerin Efektif Kullanımı
Kurumsal Ağlarda Saldırı Tespiti Amaçlı Honeypot Sistemlerin Efektif Kullanımı
 
Gerçek Dünyadan Siber Saldırı Örnekleri
Gerçek Dünyadan Siber Saldırı Örnekleri Gerçek Dünyadan Siber Saldırı Örnekleri
Gerçek Dünyadan Siber Saldırı Örnekleri
 
Czy mogą się udawać projekty publiczne?
Czy mogą się udawać projekty publiczne?Czy mogą się udawać projekty publiczne?
Czy mogą się udawać projekty publiczne?
 
DATCC Solution Matrix
DATCC Solution MatrixDATCC Solution Matrix
DATCC Solution Matrix
 
CES 2016 – GfK smart home presentation
CES 2016 – GfK smart home presentationCES 2016 – GfK smart home presentation
CES 2016 – GfK smart home presentation
 
Mobile Application Penetration Testing
Mobile Application Penetration TestingMobile Application Penetration Testing
Mobile Application Penetration Testing
 
Track 4 session 5 - st dev con 2016 - simplifying the setup and use of iot ...
Track 4   session 5 - st dev con 2016 - simplifying the setup and use of iot ...Track 4   session 5 - st dev con 2016 - simplifying the setup and use of iot ...
Track 4 session 5 - st dev con 2016 - simplifying the setup and use of iot ...
 
LLMNR ve NetBIOS Poisoning
LLMNR ve NetBIOS PoisoningLLMNR ve NetBIOS Poisoning
LLMNR ve NetBIOS Poisoning
 

Similar to Mesos Con 2016 Optimistic Offer

071310 sun d_0930_feldman_stephen
071310 sun d_0930_feldman_stephen071310 sun d_0930_feldman_stephen
071310 sun d_0930_feldman_stephenSteve Feldman
 
Dish Supply Chain: Journey to Collaborative Planning
Dish Supply Chain: Journey to Collaborative PlanningDish Supply Chain: Journey to Collaborative Planning
Dish Supply Chain: Journey to Collaborative Planningrnaramore
 
Efficient cluster resource management by using Cook and Mesos / Li Jin (Two S...
Efficient cluster resource management by using Cook and Mesos / Li Jin (Two S...Efficient cluster resource management by using Cook and Mesos / Li Jin (Two S...
Efficient cluster resource management by using Cook and Mesos / Li Jin (Two S...Ontico
 
Agile Project Outsourcing - Dealing with RFP and RFI
Agile Project Outsourcing - Dealing with RFP and RFIAgile Project Outsourcing - Dealing with RFP and RFI
Agile Project Outsourcing - Dealing with RFP and RFISiddhi
 
Agile india2012 - Dealing with agile RFP
Agile india2012   - Dealing with agile RFPAgile india2012   - Dealing with agile RFP
Agile india2012 - Dealing with agile RFPRavi Kumar
 
Modern Perspectives on Recommender Systems and their Applications in Mendeley
Modern Perspectives on Recommender Systems and their Applications in MendeleyModern Perspectives on Recommender Systems and their Applications in Mendeley
Modern Perspectives on Recommender Systems and their Applications in MendeleyMaya Hristakeva
 
Tag based recommender system
Tag based recommender systemTag based recommender system
Tag based recommender systemKaren Li
 
Podila mesos con-northamerica_sep2017
Podila mesos con-northamerica_sep2017Podila mesos con-northamerica_sep2017
Podila mesos con-northamerica_sep2017Sharma Podila
 
Cloud Ready Apps
Cloud Ready AppsCloud Ready Apps
Cloud Ready AppsDotitude
 
Multiple Models with Multiple Perspectives in a Cross-Functional Team - KanDD...
Multiple Models with Multiple Perspectives in a Cross-Functional Team - KanDD...Multiple Models with Multiple Perspectives in a Cross-Functional Team - KanDD...
Multiple Models with Multiple Perspectives in a Cross-Functional Team - KanDD...Mufrid Krilic
 
Winter is Coming: Who Will Win The Container Wars (webinar)
Winter is Coming: Who Will Win The Container Wars (webinar)  Winter is Coming: Who Will Win The Container Wars (webinar)
Winter is Coming: Who Will Win The Container Wars (webinar) The {code} Team
 
ResourceSync: Conceptual and Technical Problem Perspective
ResourceSync: Conceptual and Technical Problem PerspectiveResourceSync: Conceptual and Technical Problem Perspective
ResourceSync: Conceptual and Technical Problem PerspectiveHerbert Van de Sompel
 
How Lucene Powers the LinkedIn Segmentation and Targeting Platform
How Lucene Powers the LinkedIn Segmentation and Targeting PlatformHow Lucene Powers the LinkedIn Segmentation and Targeting Platform
How Lucene Powers the LinkedIn Segmentation and Targeting PlatformHien Luu
 
Hi Maturity in the CMMI Services Context
Hi Maturity in the CMMI Services Context Hi Maturity in the CMMI Services Context
Hi Maturity in the CMMI Services Context SPIN Chennai
 
Softchoice Discovery Series: Cloud Cost Governance
Softchoice Discovery Series: Cloud Cost GovernanceSoftchoice Discovery Series: Cloud Cost Governance
Softchoice Discovery Series: Cloud Cost GovernanceSoftchoice Corporation
 

Similar to Mesos Con 2016 Optimistic Offer (20)

071310 sun d_0930_feldman_stephen
071310 sun d_0930_feldman_stephen071310 sun d_0930_feldman_stephen
071310 sun d_0930_feldman_stephen
 
Introduction to mesos
Introduction to mesosIntroduction to mesos
Introduction to mesos
 
Real-Time Design Patterns
Real-Time Design PatternsReal-Time Design Patterns
Real-Time Design Patterns
 
Dish Supply Chain: Journey to Collaborative Planning
Dish Supply Chain: Journey to Collaborative PlanningDish Supply Chain: Journey to Collaborative Planning
Dish Supply Chain: Journey to Collaborative Planning
 
Comparison between Cloud Mirror, Mesos Cluster, and Google Omega
Comparison between Cloud Mirror, Mesos Cluster, and Google OmegaComparison between Cloud Mirror, Mesos Cluster, and Google Omega
Comparison between Cloud Mirror, Mesos Cluster, and Google Omega
 
Efficient cluster resource management by using Cook and Mesos / Li Jin (Two S...
Efficient cluster resource management by using Cook and Mesos / Li Jin (Two S...Efficient cluster resource management by using Cook and Mesos / Li Jin (Two S...
Efficient cluster resource management by using Cook and Mesos / Li Jin (Two S...
 
Agile Project Outsourcing - Dealing with RFP and RFI
Agile Project Outsourcing - Dealing with RFP and RFIAgile Project Outsourcing - Dealing with RFP and RFI
Agile Project Outsourcing - Dealing with RFP and RFI
 
Agile india2012 - Dealing with agile RFP
Agile india2012   - Dealing with agile RFPAgile india2012   - Dealing with agile RFP
Agile india2012 - Dealing with agile RFP
 
Modern Perspectives on Recommender Systems and their Applications in Mendeley
Modern Perspectives on Recommender Systems and their Applications in MendeleyModern Perspectives on Recommender Systems and their Applications in Mendeley
Modern Perspectives on Recommender Systems and their Applications in Mendeley
 
Tag based recommender system
Tag based recommender systemTag based recommender system
Tag based recommender system
 
Podila mesos con-northamerica_sep2017
Podila mesos con-northamerica_sep2017Podila mesos con-northamerica_sep2017
Podila mesos con-northamerica_sep2017
 
Cloud Ready Apps
Cloud Ready AppsCloud Ready Apps
Cloud Ready Apps
 
DevOps 101
DevOps 101DevOps 101
DevOps 101
 
Multiple Models with Multiple Perspectives in a Cross-Functional Team - KanDD...
Multiple Models with Multiple Perspectives in a Cross-Functional Team - KanDD...Multiple Models with Multiple Perspectives in a Cross-Functional Team - KanDD...
Multiple Models with Multiple Perspectives in a Cross-Functional Team - KanDD...
 
Winter is Coming: Who Will Win The Container Wars (webinar)
Winter is Coming: Who Will Win The Container Wars (webinar)  Winter is Coming: Who Will Win The Container Wars (webinar)
Winter is Coming: Who Will Win The Container Wars (webinar)
 
ResourceSync: Conceptual and Technical Problem Perspective
ResourceSync: Conceptual and Technical Problem PerspectiveResourceSync: Conceptual and Technical Problem Perspective
ResourceSync: Conceptual and Technical Problem Perspective
 
How Lucene Powers the LinkedIn Segmentation and Targeting Platform
How Lucene Powers the LinkedIn Segmentation and Targeting PlatformHow Lucene Powers the LinkedIn Segmentation and Targeting Platform
How Lucene Powers the LinkedIn Segmentation and Targeting Platform
 
ResourceSync
ResourceSyncResourceSync
ResourceSync
 
Hi Maturity in the CMMI Services Context
Hi Maturity in the CMMI Services Context Hi Maturity in the CMMI Services Context
Hi Maturity in the CMMI Services Context
 
Softchoice Discovery Series: Cloud Cost Governance
Softchoice Discovery Series: Cloud Cost GovernanceSoftchoice Discovery Series: Cloud Cost Governance
Softchoice Discovery Series: Cloud Cost Governance
 

More from Yong Feng

Client Deployment of IBM Cloud Private (Think 2019 Session 5964A)
Client Deployment of IBM Cloud Private (Think 2019 Session 5964A)Client Deployment of IBM Cloud Private (Think 2019 Session 5964A)
Client Deployment of IBM Cloud Private (Think 2019 Session 5964A)Yong Feng
 
ISTIO Deep Dive
ISTIO Deep DiveISTIO Deep Dive
ISTIO Deep DiveYong Feng
 
Cloud nativecomputingtechnologysupportinghpc cognitiveworkflows
Cloud nativecomputingtechnologysupportinghpc cognitiveworkflowsCloud nativecomputingtechnologysupportinghpc cognitiveworkflows
Cloud nativecomputingtechnologysupportinghpc cognitiveworkflowsYong Feng
 
When HPC meet ML/DL: Manage HPC Data Center with Kubernetes
When HPC meet ML/DL: Manage HPC Data Center with KubernetesWhen HPC meet ML/DL: Manage HPC Data Center with Kubernetes
When HPC meet ML/DL: Manage HPC Data Center with KubernetesYong Feng
 
Kubernetes on EGO : Bringing enterprise resource management and scheduling to...
Kubernetes on EGO : Bringing enterprise resource management and scheduling to...Kubernetes on EGO : Bringing enterprise resource management and scheduling to...
Kubernetes on EGO : Bringing enterprise resource management and scheduling to...Yong Feng
 
Edge 2016 Session 1886 Building your own docker container cloud on ibm power...
Edge 2016 Session 1886  Building your own docker container cloud on ibm power...Edge 2016 Session 1886  Building your own docker container cloud on ibm power...
Edge 2016 Session 1886 Building your own docker container cloud on ibm power...Yong Feng
 
State of Resource Management in Big Data
State of Resource Management in Big DataState of Resource Management in Big Data
State of Resource Management in Big DataYong Feng
 
IBM Platform Computing Products Connector for Apache Mesos
IBM Platform Computing Products Connector for Apache MesosIBM Platform Computing Products Connector for Apache Mesos
IBM Platform Computing Products Connector for Apache MesosYong Feng
 

More from Yong Feng (8)

Client Deployment of IBM Cloud Private (Think 2019 Session 5964A)
Client Deployment of IBM Cloud Private (Think 2019 Session 5964A)Client Deployment of IBM Cloud Private (Think 2019 Session 5964A)
Client Deployment of IBM Cloud Private (Think 2019 Session 5964A)
 
ISTIO Deep Dive
ISTIO Deep DiveISTIO Deep Dive
ISTIO Deep Dive
 
Cloud nativecomputingtechnologysupportinghpc cognitiveworkflows
Cloud nativecomputingtechnologysupportinghpc cognitiveworkflowsCloud nativecomputingtechnologysupportinghpc cognitiveworkflows
Cloud nativecomputingtechnologysupportinghpc cognitiveworkflows
 
When HPC meet ML/DL: Manage HPC Data Center with Kubernetes
When HPC meet ML/DL: Manage HPC Data Center with KubernetesWhen HPC meet ML/DL: Manage HPC Data Center with Kubernetes
When HPC meet ML/DL: Manage HPC Data Center with Kubernetes
 
Kubernetes on EGO : Bringing enterprise resource management and scheduling to...
Kubernetes on EGO : Bringing enterprise resource management and scheduling to...Kubernetes on EGO : Bringing enterprise resource management and scheduling to...
Kubernetes on EGO : Bringing enterprise resource management and scheduling to...
 
Edge 2016 Session 1886 Building your own docker container cloud on ibm power...
Edge 2016 Session 1886  Building your own docker container cloud on ibm power...Edge 2016 Session 1886  Building your own docker container cloud on ibm power...
Edge 2016 Session 1886 Building your own docker container cloud on ibm power...
 
State of Resource Management in Big Data
State of Resource Management in Big DataState of Resource Management in Big Data
State of Resource Management in Big Data
 
IBM Platform Computing Products Connector for Apache Mesos
IBM Platform Computing Products Connector for Apache MesosIBM Platform Computing Products Connector for Apache Mesos
IBM Platform Computing Products Connector for Apache Mesos
 

Recently uploaded

VIP High Class Call Girls Bikaner Anushka 8250192130 Independent Escort Servi...
VIP High Class Call Girls Bikaner Anushka 8250192130 Independent Escort Servi...VIP High Class Call Girls Bikaner Anushka 8250192130 Independent Escort Servi...
VIP High Class Call Girls Bikaner Anushka 8250192130 Independent Escort Servi...Suhani Kapoor
 
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /WhatsappsBeautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsappssapnasaifi408
 
Industrialised data - the key to AI success.pdf
Industrialised data - the key to AI success.pdfIndustrialised data - the key to AI success.pdf
Industrialised data - the key to AI success.pdfLars Albertsson
 
Dubai Call Girls Wifey O52&786472 Call Girls Dubai
Dubai Call Girls Wifey O52&786472 Call Girls DubaiDubai Call Girls Wifey O52&786472 Call Girls Dubai
Dubai Call Girls Wifey O52&786472 Call Girls Dubaihf8803863
 
Aminabad Call Girl Agent 9548273370 , Call Girls Service Lucknow
Aminabad Call Girl Agent 9548273370 , Call Girls Service LucknowAminabad Call Girl Agent 9548273370 , Call Girls Service Lucknow
Aminabad Call Girl Agent 9548273370 , Call Girls Service Lucknowmakika9823
 
Delhi Call Girls CP 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
Delhi Call Girls CP 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip CallDelhi Call Girls CP 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
Delhi Call Girls CP 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Callshivangimorya083
 
VIP High Class Call Girls Jamshedpur Anushka 8250192130 Independent Escort Se...
VIP High Class Call Girls Jamshedpur Anushka 8250192130 Independent Escort Se...VIP High Class Call Girls Jamshedpur Anushka 8250192130 Independent Escort Se...
VIP High Class Call Girls Jamshedpur Anushka 8250192130 Independent Escort Se...Suhani Kapoor
 
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service Bhilai
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service BhilaiLow Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service Bhilai
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service BhilaiSuhani Kapoor
 
Invezz.com - Grow your wealth with trading signals
Invezz.com - Grow your wealth with trading signalsInvezz.com - Grow your wealth with trading signals
Invezz.com - Grow your wealth with trading signalsInvezz1
 
RA-11058_IRR-COMPRESS Do 198 series of 1998
RA-11058_IRR-COMPRESS Do 198 series of 1998RA-11058_IRR-COMPRESS Do 198 series of 1998
RA-11058_IRR-COMPRESS Do 198 series of 1998YohFuh
 
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM TRACKING WITH GOOGLE ANALYTICS.pptx
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM  TRACKING WITH GOOGLE ANALYTICS.pptxEMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM  TRACKING WITH GOOGLE ANALYTICS.pptx
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM TRACKING WITH GOOGLE ANALYTICS.pptxthyngster
 
Predicting Employee Churn: A Data-Driven Approach Project Presentation
Predicting Employee Churn: A Data-Driven Approach Project PresentationPredicting Employee Churn: A Data-Driven Approach Project Presentation
Predicting Employee Churn: A Data-Driven Approach Project PresentationBoston Institute of Analytics
 
Delhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
Delhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip CallDelhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
Delhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Callshivangimorya083
 
High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...
High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...
High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...soniya singh
 
Schema on read is obsolete. Welcome metaprogramming..pdf
Schema on read is obsolete. Welcome metaprogramming..pdfSchema on read is obsolete. Welcome metaprogramming..pdf
Schema on read is obsolete. Welcome metaprogramming..pdfLars Albertsson
 

Recently uploaded (20)

VIP High Class Call Girls Bikaner Anushka 8250192130 Independent Escort Servi...
VIP High Class Call Girls Bikaner Anushka 8250192130 Independent Escort Servi...VIP High Class Call Girls Bikaner Anushka 8250192130 Independent Escort Servi...
VIP High Class Call Girls Bikaner Anushka 8250192130 Independent Escort Servi...
 
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
 
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /WhatsappsBeautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
 
Industrialised data - the key to AI success.pdf
Industrialised data - the key to AI success.pdfIndustrialised data - the key to AI success.pdf
Industrialised data - the key to AI success.pdf
 
Dubai Call Girls Wifey O52&786472 Call Girls Dubai
Dubai Call Girls Wifey O52&786472 Call Girls DubaiDubai Call Girls Wifey O52&786472 Call Girls Dubai
Dubai Call Girls Wifey O52&786472 Call Girls Dubai
 
Aminabad Call Girl Agent 9548273370 , Call Girls Service Lucknow
Aminabad Call Girl Agent 9548273370 , Call Girls Service LucknowAminabad Call Girl Agent 9548273370 , Call Girls Service Lucknow
Aminabad Call Girl Agent 9548273370 , Call Girls Service Lucknow
 
E-Commerce Order PredictionShraddha Kamble.pptx
E-Commerce Order PredictionShraddha Kamble.pptxE-Commerce Order PredictionShraddha Kamble.pptx
E-Commerce Order PredictionShraddha Kamble.pptx
 
Decoding Loan Approval: Predictive Modeling in Action
Decoding Loan Approval: Predictive Modeling in ActionDecoding Loan Approval: Predictive Modeling in Action
Decoding Loan Approval: Predictive Modeling in Action
 
Delhi Call Girls CP 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
Delhi Call Girls CP 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip CallDelhi Call Girls CP 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
Delhi Call Girls CP 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
 
VIP High Class Call Girls Jamshedpur Anushka 8250192130 Independent Escort Se...
VIP High Class Call Girls Jamshedpur Anushka 8250192130 Independent Escort Se...VIP High Class Call Girls Jamshedpur Anushka 8250192130 Independent Escort Se...
VIP High Class Call Girls Jamshedpur Anushka 8250192130 Independent Escort Se...
 
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service Bhilai
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service BhilaiLow Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service Bhilai
Low Rate Call Girls Bhilai Anika 8250192130 Independent Escort Service Bhilai
 
Invezz.com - Grow your wealth with trading signals
Invezz.com - Grow your wealth with trading signalsInvezz.com - Grow your wealth with trading signals
Invezz.com - Grow your wealth with trading signals
 
RA-11058_IRR-COMPRESS Do 198 series of 1998
RA-11058_IRR-COMPRESS Do 198 series of 1998RA-11058_IRR-COMPRESS Do 198 series of 1998
RA-11058_IRR-COMPRESS Do 198 series of 1998
 
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM TRACKING WITH GOOGLE ANALYTICS.pptx
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM  TRACKING WITH GOOGLE ANALYTICS.pptxEMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM  TRACKING WITH GOOGLE ANALYTICS.pptx
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM TRACKING WITH GOOGLE ANALYTICS.pptx
 
Predicting Employee Churn: A Data-Driven Approach Project Presentation
Predicting Employee Churn: A Data-Driven Approach Project PresentationPredicting Employee Churn: A Data-Driven Approach Project Presentation
Predicting Employee Churn: A Data-Driven Approach Project Presentation
 
Delhi 99530 vip 56974 Genuine Escort Service Call Girls in Kishangarh
Delhi 99530 vip 56974 Genuine Escort Service Call Girls in  KishangarhDelhi 99530 vip 56974 Genuine Escort Service Call Girls in  Kishangarh
Delhi 99530 vip 56974 Genuine Escort Service Call Girls in Kishangarh
 
VIP Call Girls Service Charbagh { Lucknow Call Girls Service 9548273370 } Boo...
VIP Call Girls Service Charbagh { Lucknow Call Girls Service 9548273370 } Boo...VIP Call Girls Service Charbagh { Lucknow Call Girls Service 9548273370 } Boo...
VIP Call Girls Service Charbagh { Lucknow Call Girls Service 9548273370 } Boo...
 
Delhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
Delhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip CallDelhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
Delhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
 
High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...
High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...
High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...
 
Schema on read is obsolete. Welcome metaprogramming..pdf
Schema on read is obsolete. Welcome metaprogramming..pdfSchema on read is obsolete. Welcome metaprogramming..pdf
Schema on read is obsolete. Welcome metaprogramming..pdf
 

Mesos Con 2016 Optimistic Offer

  • 1. Optimistic Offer - What does it mean to Mesos Frameworks? Yong Feng MesosCon North America 2016, 6/01/2016 1 yongfeng@ca.ibm.com
  • 2. Overview • The Pessimistic Offer Programming Model • Implementing frameworks to use Pessimistic Offer • Observed limitations and proposed mitigations • The Optimistic Offer Programming Model • Benefits • Design Plan • Development status • Optimistic vs Pessimistic Offer • Discussion and Additional Resources 2
  • 3. The Pessimistic Offer Programming Model • Offer: allocate available resources in a host to a single framework. Allocated resources are locked (cached) for a period of time. • Resources are allocated or partitioned among frameworks without any knowledge of a framework’s requirements • Resources in an offer is either consumed by launching task, or freed by rejecting or rescinding of an offer • Frameworks cannot consume or reject parts of an offer while still keeping the left over resources of that offer • Resources are only available for re-allocation after the offer is rejected/rescinded or the task/executor is finished • Inverse Offer: deallocate (or reclaim) resources from a framework 3 Free Allocated Used Framework launch task Reject or rescind Offer to framework Task or executor finished
  • 4. Framework Implementation: Cache Offers 4 • What if a framework has no suitable demand when an offer is received? • They cache it ! • Caching resources can lead to: • Better performance for future demand • Better scheduling decision with more candidate hosts, for example affinity … • Stockpiled resources for future tasks that have higher demands • Still want to be a good citizen? • Allow for TTL (expiration) of cached offers • Frameworks that support TTL of cached offers: • Swarm • Kubernetes
  • 5. Framework Implementation: Revive Offers 5 • What if there is no demand to consume the offers in a framework? • Reject it ! (or send suppress request) • Rejecting or suppressing unwanted resources allow for: • Better resource utilization by allocating resources to frameworks that have a demand • What if there is a demand in a framework, but no offer? • Revive the offer • Send a request to Mesos to reallocate the offer • Frameworks that support Suppress/Revive offer: • Marathon
  • 6. Pessimistic Offer: Limitations Inefficient resource allocation Mitigation Cached resource are not released to other frameworks in time Over a period of time, each offer only includes small pieces of each resource in a host. To fix this, offer TTL must be enforced by Mesos and Mesos frameworks. PO – Pessimistic Offer OO – Optimistic Offer Average time to receive offers (ms) for multiple client frameworks (Pessimistic vs Optimistic Offer) 6
  • 7. Pessimistic Offer: Limitations (Cont’d) Poor scheduling quality Mitigation Offered resource may not be the most suitable resource Frameworks should delay the scheduling decision until they get suitable resources PO – Pessimistic Offer OO – Optimistic Offer Job duration time (ms) for multiple client frameworks (Pessimistic vs Optimistic Offers) 7
  • 8. Pessimistic Offer: Limitations (Cont’d) Low QoS guarantee Mitigation Long tasks prevent resource re-allocation. Hard to preempt resource among frameworks. Implement Quotas, reservations, over- subscription … https://www.youtube.com/watch?v=jC8lhGQN2Sc 8
  • 9. Can We Squeeze More Out of Pessimistic Offers? 9 • Adjust the size of an offer (Mesos-3765) • Smaller sized offers improves resource utilization, and fairness • Oversubscription for reservation (Mesos-4967) • Lend reserved resource to other framework if it is not being used. This improves resource utilization without impacting QoS • Reuse the recovered resource without delay (Mesos-3078 or Mesos-4811) • Accelerate scheduling so that resources can be used once it becomes available • More hints in filter when rejecting offer • Avoid ping-pong of resources between Mesos and framework to improve performance and avoid starvation
  • 10. The Optimistic Offer Model Offer: displays all available resources to multiple frameworks.
  • 11. Benefits of the Optimistic Offer Model • Better resource utilization • Available resources are always visible to all frameworks • Improved scheduling performance • Resources recovered from finished tasks are visible to all frameworks immediately • Enhanced quality of scheduling decision • Increased availability of resources help frameworks to make better scheduling decisions • QoS guarantee • QoS of workload is enforced by pre-emption 11
  • 12. The Optimistic Offer - Design Plan • Track both offered resources and consumed resources • Offered resources are not allocated resources. Offered resources are viewed as available resources to all framework • Consumed resources are equivalent to allocated resource in the Pessimistic Offer model, however they might still be viewed as available resources to some frameworks under special conditions. • Offered resources are based on resource plan or usage instead of allocation • Available resources are visible to frameworks via Offer • Available resources are decided by resource plan and resource usage • Refresh resource and quota availability in a timely manner • Update resource availability and status to framework by either rejecting offer from framework or rescinding offer from master • Update resource plan availability with quota or restrictions to each framework • Pre-empt tasks to enforce the QoS • Pre-empt task from lower priority framework or overused framework to enforce QoS • Export the resource status to framework to achieve “smart” pre-emption 12
  • 13. The Optimistic Offer - Design Plan (Cont’d) • Programming model: • Allow frameworks to use offers without the need for rejection of the offer from a prior framework • Refresh the status of a resource by handling rescind messages or rejected offers • Use the non-revocable resource first and then revocable resource. • Handle the inverse offer smartly for pre-emption 13
  • 14. The Optimistic Offer - Development Status • Mesos-1607 Phase 1 (renamed as “Oversubscription for reservation” in Mesos-4967) • Target: ~4Q 2016 • Status: Reviewable; ~30 patches are ready for review, try it after applying the patches 14 Total Resources Reserved Allocated Used Revocable
  • 15. The Optimistic Offer - Development Status (Cont’d) 15 • Mesos-1607 Prototype • Target: IBM Mesos Connector • Design: adjust allocator API to track consumed resources and resolve resource races • Status: In Bluemix Container Cloud Service, and planned in Mesos roadmap.
  • 16. Optimistic Offer vs Pessimistic Offer Model 16 Idle frameworks Idle frameworks with smaller tasks PO – Pessimistic Offer OO – Optimistic Offer (Number of task per ms)
  • 17. Optimistic Offer vs Pessimistic Offer Model 17 PO – Pessimistic Offer OO – Optimistic Offer (Number of task per ms) Idle frameworks Idle frameworks with smaller tasks and small TTL
  • 18. Optimistic Offer vs Pessimistic Offer Model 18 PO – Pessimistic Offer OO – Optimistic Offer (Number of task per ms) Idle frameworks Idle frameworks when placement constraint is added
  • 19. Optimistic Offer vs Pessimistic Offer Model 19 PO – Pessimistic Offer OO – Optimistic Offer (Job duration per ms for multiple clients) Idle frameworks Idle frameworks with smaller tasks
  • 20. Optimistic Offer vs Pessimistic Offer Model 20 PO – Pessimistic Offer OO – Optimistic Offer (Job duration per ms for multiple clients) Idle frameworks Idle frameworks with smaller tasks and smaller TTL
  • 21. Optimistic Offer vs Pessimistic Offer Model 21 PO – Pessimistic Offer OO – Optimistic Offer (slave utilization) Idle frameworks Idle frameworks with smaller tasks
  • 22. Optimistic Offer vs Pessimistic Offer Model 22 PO – Pessimistic Offer OO – Optimistic Offer (slave utilization) Idle frameworks Idle frameworks with smaller tasks and smaller TTL
  • 23. Evaluation – Conflicts • 4 swarm frameworks • 10 nodes, 1000 containers (tasks) • Task request rejected after 10 placement retries (a) System load = 0.8 and varied request resource size (b) System load = 0.4,0.8,0.99 and random request resource size 0 5 10 15 20 25 small medium large varied Percentageoftasks Task resource request size Conflicted tasks (%) Rejected tasks (%) 0 2 4 6 8 10 12 14 16 40% 80% 99% Percentageoftasks System Load Conflicted tasks (%) Rejected tasks (%)
  • 24. Optimistic Offer in IBM Bluemix Container Cloud 24 Mesos Master Swarm Scheduler Docker CLI/API framework Kubernetes Scheduler Mesos Agents Offers framework Offers Tasks to Mesos Kubernetes CLI/API Mesos Agents Mesos Agents Network Agents Network Agents Tasks to Mesos
  • 25. Open Questions • How to export more policy status and configuration to frameworks so that frameworks can make better decisions? • How to reduce scheduling overheads, such as resource conflicts? • … 25
  • 26. IBM Booth 26 Welcome to S5 • Spark benchmark with IBM Session Scheduler • Bluemix Container Service • GPU in Power • Mesos Connector
  • 27. Mesos Community Activities 27 • Active development with Mesos community – 11 IBM Developers. • 100+ JIRAs delivered or in progress • Leading or participating in several work streams: POWER Support, Optimistic Offers, Container Support, GPU Support, Swarm and Kubernetes integration • Relationship with Mesosphere – weekly calls, on-site developer presence • Attendance at MesosCon 2016 with sponsorship and booth • Aligning with IBM Container Service to leverage common OSS technologies • Technical Preview of Mesos with IBM Value-Add (ASC) on Docker Hub – Both x86 and POWER images
  • 28. IBM Committed Mesos Patches 28 For the period June 2015 to May 2016