Platform Computing

IBM Cloud Service Ready to use Platform LSF & clusters in the SoftLayer cloud

February 25, 2014

1 © 2014 IBM Corporation Platform Computing Agenda

v Mapping clients needs to cloud technologies v Addressing your pain points v Introducing IBM Platform Computing Cloud Service v Product features and benefits v Use cases v Performance benchmarks

2 © 2014 IBM Corporation Platform Computing HPC cloud characteristics and economics are different than general-purpose computing

• High-end hardware and special purpose devices (e.g. GPUs) are typically used to supply the needed processing, memory, network, and storage capabilities • The performance requirements of technical computing and service-oriented workloads means that performance may be impacted in a virtualized cloud environment, especially when latency or I/O is a constraint • HPC cluster/grid utilization is usually in the 70-90% range, removing a major potential advantage of a public cloud service provider for stable workload volumes

HPC Workloads Recommended for Private Cloud

HPC Workloads with Best Potential for Virtualized Public & Hybrid Cloud

Primary HPC Workloads

3 © 2014 IBM Corporation Platform Computing IBM’s HPC cloud strategy provides a flexible approach to address a variety of client needs

Private Hybrid Public Clouds Clouds Clouds

Evolve existing infrastructure to Enable integrated HPC Cloud to enhance approach to improve Access additional responsiveness, HPC cost and HPC capacity with flexibility, and capability variable cost model cost effectiveness. 60%

Based on HPC Cloud’s potential impact, organizations are evolving their infrastructures to enable private cloud deployments, exploring hybrid clouds, and considering public clouds.

4 © 2014 IBM Corporation Platform Computing Are you experiencing any of these pain points?

• Unable to meet business objectives (delay to market, etc.) • Existing resources insufficient to meet peek compute demand – Long run times on existing cluster or grid – No access to local technical computing resources (workstation users) • Technical resources expensive and time consuming to acquire • The skills/staff to architect and manage a technical computing infrastructure can be difficult to acquire

Planned Daily Cycle (24 x 365) Planned Project

50,000 1600 1400 40,000 1200 30,000 1000 800 20,000 600 400 10,000 200 - 0 1 4 7 10 13 16 19 22 April May June Financial Services Life Sciences

5 © 2014 IBM Corporation Platform Computing IBM Platform Computing Cloud Service Making the cloud work for you

Build Manage Support Protect • Complete, ready to run • Seamless workload • 24X7 cloud operation • Data encryption, clusters in the cloud management, on- support dedicated physical • Add additional capacity premise and in the • Access to technical machines and network in hours instead of cloud computing expertise • Security through months • Transparent user when you need it physical isolation experience

Complete, end to end dynamic cloud solution

6 © 2014 IBM Corporation Platform Computing Ready to use Platform LSF & Platform Symphony clusters in the cloud

Client and ISV Applications

IBM Platform Computing Cloud Service (SaaS)

IBM Platform IBM Platform LSF Symphony

SoftLayer, an IBM Company Infrastructure

24X7 CloudOps Support

7 © 2014 IBM Corporation Platform Computing Dedicated physical and virtual machine infrastructure as a service

• 13+ data centers • 17 network PoPs 190,000+ 21,000+ 22,000,000+ SERVERS CUSTOMERS DOMAINS • Global private network • Bare metal and virtual machines

8 © 2014 IBM Corporation Platform Computing Ready to use Platform LSF & Platform Symphony clusters in the cloud

DIFFERENTIATOR RATING IBM ADVANTAGES

• SoftLayer’s architecture outperforms by >50% equivalent Workload I/O intensity Low High intensity AWS instances for high I/O intensity workloads workloads workloads

• SoftLayer offers hundreds of Control (APIs, hardware configurations vs. 14 hardware / network Low degree of High degree of for AWS configurability) control and control and • ~2,000 APIs for SoftLayer vs. ~60 customization customization for AWS and none for RAX

• Unified integration & control panel for multiple cloud Integrated platform of architectures multiple architectures Single platform Seamless integration • RAX requires paid bridge, different control interfaces

AWS RAX IBM

9 © 2014 IBM Corporation Platform Computing Non-shared physical machines for added security and performance

• Dedicated and isolated compute environment • All machine instances are dedicated to the client • Each cluster is isolated on a VLAN • Only the VPN gateway has an addressable interface • All customer data at rest is encrypted on shared file systems • When machines instances are decommissioned the disks are scrubbed using DoD approved methods

10 © 2014 IBM Corporation Platform Computing Optimal performance for technical computing apps EDA Benchmark (IBM-MESA)

Industrial Manufacturing Benchmark – Structural Mechanics

Note: Benchmark results were obtained by IBM and have not yet been externally 11 audited or validated. © 2014 IBM Corporation Platform Computing Run and supported by dedicated, 24X7 HPC Cloud Operations Team

CloudOps functions • Pre-provisioning: Provide guidance to client on how to enable VPN, multi-cluster settings & security settings on the client on-premise environment • One time setup testing: Extensive testing of the cluster prior to release to the client • Extensive testing of the cluster on every event of flex-up prior to release to the client • Email alerts prior to flex-down & cluster shutdown operations • Email alerts in case of any overage (compute hours, download bandwidth) • Provide billing details of monthly usage including overage details • Provide support under IBM SLA by experts highly experienced in Platform Computing products

Value: quality, peace of mind & minimum disruption to business • Extensive quality checks ensures minimum loss of usage hours & disruptions • Proactive alerts ensures that in-progress critical jobs are not killed in case of Flex-down & Cluster Shutdowns and Overages • Highly trained & experienced Support ensures smooth on-boarding and minimize disruptions

12 © 2014 IBM Corporation

Platform Computing Industry-leading workload management

• 20 years managing distributed scale-out systems with 2000+ 23 of 30 customers in many industries largest commercial • High performance workload management combined with enterprises intelligent resource scheduling engine 60% of top • Unmatched scalability (small clusters to global grids) and financial production-proven reliability services companies • Heterogeneous – manages System x and Power plus 3rd party systems, virtual and bare metal, accelerators / GPU, cloud, etc. Over 5M • Shared services for both compute and data intensive workloads CPUs under management • Integrated solutions with vertical reference architectures

13 © 2014 IBM Corporation Platform Computing IBM Platform LSF Overview Powerful workload management for demanding, distributed and mission-critical high performance computing environments. Key Capabilities • Powerful - Policy and resource-aware scheduling - Resource consolidation for optimal performance - Advanced self-management • Flexible - Heterogeneous platform support - Policy-driven automation - CLI, web services, APIs • Scalable - Thousands of concurrent users and jobs - Virtualized pool of shared resources - Flexible control, multiple policies Client Benefits • Optimal utilization: reduced infrastructure cost • Robust capabilities: improved productivity • High throughput: faster time to results 14 14 © 2014 IBM Corporation Platform Computing IBM Platform Symphony

Overview Low-latency grid management platform for distributed computing and analytics with sophisticated resource sharing Key Capabilities • Accelerates service-oriented applications • Extreme app scalability and throughput with very low latency • Compute and data-intensive applications on a single platform • Sophisticated, hierarchical resource sharing • Open and flexible: choice of OS, frameworks and Low Latency / High throughput Sub-millisecond, 17,000 tasks per second languages Large Scale Client Benefits 10k cores per application, 40k cores per grid Efficient shared services • Increase performance and analytic result quality Heterogeneous & Open • Reduces IT costs - increase utilization, simplify , Windows, AIX, C/C++, C#, Java, Excel, application onboarding, reduce administration costs Python, R

15 15 © 2014 IBM Corporation Platform Computing Use case 1 – hybrid cluster

The problem • Existing resources cannot meet peak demand • Resources are expensive and time consuming to acquire • Skills to architect and manage clusters are difficult to find • Fixed or reduced budgets • On-premise constraints in space, cooling and power The solution • Fully functioning IBM Platform LSF or Symphony clusters are provisioned on the SoftLayer cloud and connected to the on- premise cluster, expanding capacity as needed • Leverage MultiCluster capability for managed forwarding of jobs from on premise cluster to off premise cluster The Value • Access to additional compute capacity on a temporary basis as needed • Near-zero wait times • Reduce costs by paying for only what is used • Pay for additional capacity as an operating expense • Fully supported, end-to-end solution, from the on-premise to the on-cloud clusters • Expected and reliable performance from running technical computing workloads on physical machines • Transparent access to cloud resources, the end user experience does not change 16 © 2014 IBM Corporation Platform Computing Use case 2 – stand-alone cluster in the cloud

The problem • New and emerging need for technical computing • Skills to architect and manage clusters are difficult to find • Resources are expensive and time consuming to acquire • Inconsistent demand does not justify the investment The solution • Fully functioning Platform LSF and Symphony clusters are provisioned on the SoftLayer cloud providing resources as needed

The value § Market-leading Platform LSF and Platform Symphony § Access to technical computing resources on a temporary basis without the need to acquire, install and configure the infrastructure and cluster software § Keep costs low by paying for only what is used § Pay for capacity as an operating expense § Fully supported solution § Expected and reliable performance from running workloads on physical machines

17 © 2014 IBM Corporation Platform Computing Is IBM Platform Computing Cloud Service a good fit for you?

Business pain points • And you experiencing lost profit due to missed deadlines? • Do you experience pressure to convert your compute environment capital expense to operational expense? • Have you ever missed a deadline or delayed a project because technical computing resource procurement took too long ?

Technology pain points • Do your users ever scale back their analyses to lower fidelity or less accuracy in order to fit them into the local compute environment or to a time window? • Do you regularly, occasionally, or permanently have fewer resources (CPUs, disk, memory, etc) than you would like to have to service the user’s compute demand? • Do you experience a large variance in compute resource utilization? • Have you reached, or will you reach the capacity of your datacenter(s), and do you need a plan to grow beyond that capacity ? • Are your customers asking you for cloud licenses for Platform LSF or Platform Symphony?

18 © 2014 IBM Corporation Platform Computing IBM Platform Computing Cloud Service Making the Cloud Work for You

IBM Hybrid Cloud On On Premise SmartCloud

powered by Software & Systems

Unmatched Capabilities Cloud Leadership Policy-driven Workload Expertise from Management Client Engagements

Unmatched Expertise Analytics, Technical Computing, Software, Services and ISV Partnerships

Consolidation Supporting heterogeneous IBM and non-IBM infrastructure

19 © 2014 IBM Corporation Platform Computing

Thank You

20 © 2014 IBM Corporation Platform Computing SoftLayer and Amazon EC2 Products tested

IaaS CPU Cores Memory Disk Space Physical / Hourly NAME Provider (GB) (GB) Virtual Rate (USD) SoLayer 16 64 1000[1] Physical $1.85[2] SL PM SoLayer 8 8 500[3] Virtual $0.88 SL VM SoLayer 16 64 1000[1] Physical $3.83[5] SL PM (ded) Amazon 32 60.5 3360 Virtual $2.40[4] EC2 CC2 EC2 (CC2) Amazon 8 7 840 Virtual $0.58 EC2 EC2 2XL (c1.xlarge)

SL Physical Machine Intel(R) Xeon(R) CPU E5-2650 0 @ 2.00GHz SL Physical Machine (dedicated) Intel® Xeon® CPU E5-2690 0 @ 2.90GHz SL Virtual Machine Intel(R) Xeon(R) CPU E5-2650 v2 @ 2.60GHz Amazon CCI2 Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz Amazon 2XL Intel(R) Xeon(R) CPU E5-2650 0 @ 2.00GHz

21 © 2014 IBM Corporation Platform Computing Memory Bandwidth STREAM (higher is better)

9000

8000

7000

6000 COPY 5000 SCALE 4000 ADD 3000 TRIAD 2000 STREAM Price Performance (higher is better) 1000 4,500.00 COPY 0 4,000.00 SCALE SL PM SL VM EC2 CCI2 EC2 2XL SL PM (ded) 3,500.00 ADD 3,000.00 TRIAD 2,500.00 2,000.00 1,500.00 1,000.00 500.00 0.00 SL PM SL VM EC2 CCI2 EC2 2XL SL PM (ded)

22 © 2014 IBM Corporation Platform Computing CPU Performance

SuperPI (lower is better)

800 700 600 500 400 300

Elapsed Time Time Elapsed 200 100 0 SL PM SL VM EC2 CCI2 EC2 2XL SL PM (ded) SuperPI Price-Performance (higher is better) 10.00

8.00

6.00

4.00

2.00 throughput per dollar perdollar throughput 0.00 SL PM SL VM EC2 CCI2 EC2 2XL SL PM (ded)

23 © 2014 IBM Corporation Platform Computing

Network Bandwidth openMPI 100000

10000

1000 SLVM EC2 2XL EC2 CCI2 SL PM 100

Bandwidth (Mbits/s) (Mbits/s) Bandwidth SL PM Dedicated

10

1 1 10 100 1000 10000 100000 1000000 10000000 Message Size (Bytes)

24 © 2014 IBM Corporation Platform Computing Network Latency

openMPI Latency (lower is better)

120

100

80

60

40

20

0 SL VM MPI 2 node EC2 2XL MPI 2 node EC2 CCI2 MPI 2 SL PM MPI 2 node SL PM (ded) MPI 2 node node

25 © 2014 IBM Corporation Platform Computing Input / Output Performance I/O Bandwidth - WRITE (higher is better)

350000

300000

250000 SL VM Write 200000 EC2 2XL Write

kB/sec 150000 EC2 CCI2 Write 100000 SL PM Write 50000 SL PM Ded Write 0 0 1 2 3 4 5 I/O file size (factor of memory size) I/O Bandwidth - READ (higher is better)

400000 350000 300000 250000 SL VM Read 200000 EC2 CCI2 Read

kB/sec 150000 EC2 2XL Read 100000 SL PM Read 50000 SL PM Ded Read 0 0 1 2 3 4 5 I/O file size (factor of memory size) 26 © 2014 IBM Corporation Platform Computing Software Compilation Software Compile Performance (lower is better) 800 700 600 500 400 300

Elapsed Time (s) Time Elapsed 200 100 0 SL VM SL PM EC2 2XL EC2 CCI SL PM Ded Software Compile Price-Performance (higher is better) 9.00 8.00 7.00 6.00 5.00 4.00 Runs / $ 3.00 2.00 1.00 0.00 SL VM SL PM EC2 2XL EC2 CCI SL PM Ded 27 © 2014 IBM Corporation Platform Computing Life Science (BWA) Life Sciences Benchmark (BWA) 40000 (lower is better)

35000

30000

25000

20000

15000

Elapsed time(sec)Elapsed 10000

5000

0 SL PM (ded) SL PM SL VM EC2 CCI2 EC2 2XL Series1 20846.481 26509.368 25897.44 22442.7 37491 Life Sciences Benchmark (BWA) Price

25.00 Performance (lower is better)

20.00

15.00

$ / run 10.00

5.00

0.00 SL PM (ded) SL PM SL VM EC2 CCI2 EC2 2XL 28 Series1 22.21 7.79 6.33 14.96 6.04 © 2014 IBM Corporation Platform Computing EDA Benchmark (IBM-MESA) EDA - IBM Mesa (lower is better) 3500

3000

2500

2000

1500

1000 Elapsed Time (sec) Time Elapsed 500

0 SL PM (ded) SL PM SL VM EC2 2XL EC2 CCI2 EDA - IBM Mesa - Price-Performance (higher is better) 2.50

2.00

1.50

Runs / $ 1.00

0.50

0.00 SL PM (ded) SL PM SL VM EC2 2XL EC2 CCI2

29 © 2014 IBM Corporation Platform Computing Provisioning Time

Provisioning Time (sec) (lower is better)

100000

10000

1000

100

10

1 SL PM SL VM EC2 CCI2 EC2 2XL SL PM Ded

30 © 2014 IBM Corporation Platform Computing Industrial Manufacturing – Structural Mechanics

One Node - S4D One Node - S6

13 7

11 6 5 SL PM 9 SL PM 4 EC2 CCI2 7 EC2 CCI2 SL VM SL VM 3 5 EC2 2XL EC2 2XL 2 3 SL PM (ded) SL PM (ded) 1 0 2 4 6 8 10 12 14 16 1 (relativeEC2 2XL) to Speedup Speedup (relativeEC2 2XL) to Speedup 0 2 4 6 8 10 12 14 16 CPUs CPUs Two Nodes - S6 Two Nodes - S4D 9 19 8 17 7 15 6 SL PM 13 SL PM 11 5 EC2 CCI2 EC2 CCI2 9 4 SL VM 7 SL VM 3 EC2 2XL 5 EC2 2XL 3 2 SL PM (ded) SL PM (ded) 1 (relativeEC2 2XL) to Speedup 1

Speedup (relativeEC2 2XL) to Speedup 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 0 2 4 6 8 1012 14 1618 20 22 24 2628 30 32 CPUs CPUs 31 © 2014 IBM Corporation Platform Computing Industrial Manufacturing – CFD

OpenFoam Speedup Backplane (higher is better) 18 16 14 12 SL PM (ded) 10 SL PM 8 SL VM 6 EC2 CCI2 4 EC2 2XL 2 Speedup (relativeEC2 2XL) to Speedup 0 OpenFoam Speedup Ethernet 1 3 5 7 9 11 13 15 (higher is better) # cores 8 7 6 SL PM (ded) 5 SL PM 4 SL VM 3 EC2 CCI2 2 EC2 2XL 1 Speedup (relativeEC2 2XL) to Speedup 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 # cores 32 © 2014 IBM Corporation