Rev Up Your HPC Engine

Rev Up Your HPC EngineFritz Ferstl, CTO Univa Corp, [email protected]

Who is Univa?

Copyright © 2014 Univa Corporation. All Rights Reserved.2

• Profile• Based in Chicago, global

reach

• >500 customers in 3 yrs (mostly Fortune 500)

• Products /Technologies:• Univa Grid Engine

• UniSight

• Univa License Orchestrator

• UniCloud

Data Center Automation ExpertsDo more with less in Big Compute and Big Data

Help organizations play a better game

of Tetris

Challenges for Workload and Resource Management Systems

Copyright © 2014 Univa Corporation. All Rights Reserved. 3

Scalability

• Node counts stay flat or go down, sockets stay flat, cores explode• With the core explosion, the number of jobs also explodes

• Ever shorter run-times, more applications, more use cases

• Large commercial sites approach or go beyond 100K

• Throughput clustersprocess >150 millionjobs / month

4Copyright © 2014 Univa Corporation. All Rights Reserved.

Heterogeneity


• Hardware• Multi-sockets, multi-cores

• Partial cluster upgrades

• Evolving memory, network and storage architectures

• Accelerators: GPUs, Phi

• Job Profiles• Throughput

• Array Jobs

• Large Parallel

• Interactive

• Sessions

• Reservations

• Transactional

• Hybrid

• Dependencies, Workflows

Policy Variety


• Automated Transparency?

• Manual overrides

• Preferential access

• Priorities

• Reservations

• Resource Urgencies

• Quotas

• Deadlines

• Conflict Resolution• E.g. don‘t starve large

parallel plus maintainhigh utilization

Use Case Variety


• Classical HPC (simulation) Large parallel / many mid-size parallel

• Verification / Test Throughput

• From single simulation to parameter study array jobs

• Ultra-short jobs

• Big Data / Data Mining

• Exclusive usage of nodesvs shared usage

Geographical Distribution / Clouds


• Resource sharing: servers, licenses, data, other

• Data access latencies

• Security

• File system dependencies• Pre-/Post-Staging

• Data locality:• Bring the job to the data

• Or bring the data to the job

SolutionsApproachesBest Practices


Evolve

• Architecture Evolution• more cores / nodes / jobs

make it faster

• Integration with GPUs, Phi, etc

• New Scheduling Algorithms• Efficient handling of job mixes:

parallel / array / sequential jobs

• Scheduling of ultra-short jobs

• More Monitoring, Better Error Tracking

• Reporting, Accounting & Analytics


Be Street-Smart

• Simplify where possible!

• Be-all solution can be themost expensive• Effort

• Poor utilization slow ROI

• Focus on most important goals


Think Different

• Examples:• Less HA @ more throughput via fast SSD-Raid with

regular back-up

• Use array jobs whereever possible

• More smaller jobs vs fewer biggerjobs

• All considered, preemption maybe a good option


Accept Difference

• Simple: temporarily designate parts of cluster

• Advanced: Cloud-share• Share resources across separate workload

management system instances

• Dynamically re-assign resources(servers) based on demand

• Provides autonomy whilemaintaining high utilization

• But avoid meta-schedulingwhere you can!


Tailored Solutions

• Tailoring & add-ons can make all the difference

• Tailoring such as• Job Classes

• Customized reports

• Add-ons such as• Submission portals

and wrappers


Conclusions

• Workload & Resource Management Systems more required than ever

• Specifically in the “new” era of Cloud and Big Data

• Allows you to benefit from 20+ years of experience in HPC workload orchestration and to move beyond

• Clear-cut set of challenges non-trivial solutions

• Build on best-in-class products, architectures and development teams

• Being “street-smart” about architecting and configuration of a cluster has big impact


Thank Youhttp://[email protected]


http://www.univa.com/

mailto:[email protected]

Rev Up Your HPC Engine

Technology

univa corporation

cto univa corp

jobs month

number of jobs

best practices copyright

cores nodes jobs

big data help organizations

large parallel plus