Transcript Speech title here - HPC Advisory Council

High Productivity Computing
With Windows
Windows HPC Server 2008
Agenda
What’s new in Windows HPC Server 2008
HPC Systems Management
• Administration
• Deployment
• Integration with System Center
Networking and MPI
Job Scheduling
• Service Oriented Architectures
• Job Policies
• Interoperability
Tools for Parallel Programming
Partnerships
Why Microsoft in HPC?
Current Issues
 HPC and IT data centers merging: isolated cluster
management
 Developers can’t easily program for parallelism
 Users don’t have broad access to the increase in processing
cores and data
How can Microsoft help?
 Well positioned to mainstream integration of application
parallelism
 Have already begun to enable parallelism broadly to the
developer community
 Can expand the value of HPC by integrating productivity and
management tools
Microsoft Investments in HPC
Comprehensive software portfolio: Client, Server,
Management, Development, and Collaboration
Dedicated teams focused on Cluster Computing
Unified Parallel development through the Parallel Computing
Initiative
Partnerships with the Technical Computing Institutes
Microsoft’s Productivity Vision for HPC
Windows HPC allows you to accomplish more, in less time, with reduced effort by
leveraging users existing skills and integrating with the tools they are already using.
Administrator





Integrated Turnkey HPC
Cluster Solution
Simplified Setup and
Deployment
Built-In Diagnostics
Efficient Cluster Utilization
Integrates with IT
Infrastructure and Policies
Application Developer





Integrated Tools for Parallel
Programming
Highly Productive Parallel
Programming Frameworks
Service-Oriented HPC
Applications
Support for Key HPC
Development Standards
Unix Application Migration
End - User



Seamless Integration with
Workstation Applications
Integration with Existing
Collaboration and Workflow
Solutions
Secure Job Execution and
Data Access
Windows HPC Server 2008
• Complete, integrated platform for computational clustering
• Built on top the proven Windows Server 2008 platform
• Integrated development environment
Windows Server 2008
HPC Edition
• Secure,
Reliable, Tested
• Support for high
performance hardware
(x64, high-speed
interconnects)
Microsoft HPC Pack
2008
•
•
•
•
Job Scheduler
Resource Manager
Cluster Management
Message Passing Interface
Microsoft Windows HPC
Server 2008
• Integrated Solution
out-of-the-box
• Leverages investment in
Windows administration
and tools
• Makes cluster operation
easy and secure as a
single system
Evaluation available from http://www.microsoft.com/hpc
What’s new for HPC in Windows Server 2008?
Reliability
• High Availability Services - failover now uses iSCSI
• File Systems - Self-Healing NTFS
Manageability
• Windows 64-bit Hypervisor (HyperV)
• New Event Log/Diagnostics
• PowerShell
Performance
• New TCP/IP stack with Winsock Direct improvements
• Memory Manager Performance
• Fewer, faster, and larger disk writes for pagefile and mapped file I/O (64 KB limit removed)
• I/O prioritization
• I/O priority based on the priority of the issuing thread or the explicitly set I/O priority
• Used by Windows Server 2008 background tasks like indexing and Windows Defender scans
• Numerous enhancement for Non-Uniform Memory Access(NUMA)
• SMB 2.0
• Supports NTFS client-side symbolic links
• Operations can be batched to minimize client/server round trips
• Support for arbitrary buffer sizes for more efficient copies result in 30-40x throughput
improvement
What’s new in the HPC Pack 2008?
 New System Center UI
 PowerShell for CLI Management
High Availability for Head Nodes
Windows Deployment Services
Diagnostics/Reporting
Support for Operations Manager
 Support for SOA and WCF
 Granular resource scheduling
 Improved scalability for larger
clusters
 New Job scheduling policies
Interoperability via HPC Profile
Systems
Management
Networking
& MPI
 NetworkDirect (RDMA) for MPI
 Improved Network
Configuration Wizard
 Shared Memory MS-MPI for
multi-core
 MS-MPI integrated with
Windows Event Tracing
Job
Scheduling
Storage
 Improved iSCSI SAN & parallel
file system Support in Win2008
 Improved Server Message
Block ( SMB v2)
 New 3rd party parallel system
file support for Windows
 New Memory Cache Vendors
Spring 2008, NCSA, #23
9472 cores, 68.5 TF, 77.7%
Spring 2008, Umea, #40
5376 cores, 46 TF, 85.5%
Spring 2008, Aachen, #100
2096 cores, 18.8 TF, 76.5%
Fall 2007, Microsoft, #116
2048 cores, 11.8 TF, 77.1%
30% efficiency
improvement
Windows HPC Server 2008
Spring 2007, Microsoft, #106
2048 cores, 9 TF, 58.8%
Windows Compute Cluster 2003
Spring 2006, NCSA, #130
896 cores, 4.1 TF
Winter 2005, Microsoft
4 procs, 9.46 GFlops
Windows HPC Server 2008
Ready for Prime-time
Location
Hardware – Machines
Champaign, IL
Dell blade system with 1,200
PowerEdge 1955 dual-socket, quadcore Intel Xeon 2.3 GHz processors
Hardware – Networking
#23
Summer
2008
InfiniBand and GigE
Number of Compute Nodes
Total Number of Cores
1184
9,472 cores
Total Memory
Particulars of for current Linpack
Runs
Best Linpack rating
Best cluster efficiency
For Comparison…
Linpack rating from November
2007 Top500 run (#14) on the
same hardware
Cluster efficiency from November
2007 Top500 run (#XX) on the
same hardware
Typical Top500 efficiency for
Clovertown motherboards w/ IB
regardless of Operating System
9.6 terabytes
68.5 TFPs
77.7%
68.5 TFPs
69.9%
65-77%
About 4 hours to deploy
7.8% improvement in
efficiency on the same
hardware running
Linux
Systems Management
Improved Efficiency for the Systems Admin
•
Simple to setup and manage in a familiar environment
–
–
Turnkey cluster solutions through OEMs
Simplify system and application deployment
•
•
Focus on ease of management
–
–
–
•
Comprehensive diagnostics , troubleshooting
and monitoring
Familiar, flexible and “pivotal” management
interface
Equivalent command line support for
unattended management
Scale up
–
–
–
–
•
Base images, patches, drivers, applications
Scale deployment, administration,
infrastructure
Head node failover
Cluster usage reporting
Compute node filtering
Better integration with enterprise
management
–
–
–
–
Patch Management
System Center Operations Management
PowerShell
Windows 2008 high Availability Services
Head Node High Availability
• Eliminates single point of failure with support for high availability
• Requires Windows Server 2008 Enterprise Failover Clustering Services
– Next generation of cluster services
– Major improvement in
configuration validation and
management
• HPC Pack Includes
– Setup integration with Failover
Clustering Services
• Head Node and Failover Node set
up with SQL Failover Cluster
• Job Scheduler services failover
– Management console linked to
Windows Server Failover
Management console
Private
Network
Windows
Failover
Clustered
Head node
Win2008 Enterprise
Clustered SQL
Server
Failover Head node
Win2008 Enterprise
Clustered SQL
Server
Shared
Disk
System Center Operations Manager for HPC
A more productive HPC environment
• Canned reports for end-user perspective monitoring
• Security logs analysis and reporting
Scalable Monitoring
• Monitor apps running in a scale out, distributed environment
• Scale using tiered management servers
• Agent-less Monitoring
Increased Efficiency and Control
• More secure by design
• Integration with Active Directory
• Extended solution with Management Packs
Networking & MPI
NetworkDirect
A new RDMA networking interface built for speed and stability
Priorities
– Comparable with hardware-optimized MPI
stacks
•
Socket-Based
App
Focus on MPI-Only Solution for version 2
MS-MPI
– Verbs-based design for close fit with
native, high-perf networking interfaces
– Coordinated w/ Win Networking team’s
long-term plans
•
Implementation
– MS-MPIv2 capable of 4 networking paths:
•
•
•
•
MPI App
Shared Memory
between processors on a motherboard
TCP/IP Stack (“normal” Ethernet)
Winsock Direct (and SDP)
for sockets-based RDMA
New RDMA networking interface
– HPC team partners with networking IHVs
to develop/distribute drivers for this new
interface
Windows Sockets
(Winsock + WSD)
RDMA
Networking
Networking
Networking
WinSock
Direct
Hardware
Hardware
Provider
Networking
Networking
NetworkDirect
Hardware
Hardware
Provider
Networking Hardware
Hardware
Networking
User
Mode Access Layer
TCP/Ethernet
Networking
TCP
Kernel By-Pass
•
IP
NDIS
Networking
Networking
Mini-port
Hardware
Hardware
Driver
Networking Hardware
Hardware
Networking
Hardware Driver
Networking Hardware
Hardware
Networking
Networking
Hardware
(ISV) App
CCP
Component
OS
Component
IHV
Component
User
Mode
Kernel
Mode
High Speed Networking Technologies
Bandwidth
Cisco
Voltaire
Qlogic
Open Fabrics
NetEffect
Myricom
Availability
Job Scheduling
What’s new in in Job Scheduling
Broader application support
Expanded Job Policies
Support for Job Templates
Improve interoperability
with mixed IT infrastructure
Broader Application Support
2003 (focusing on batch jobs)
2008 (focusing on Interactive jobs)
Engineering
Applications
Oil & Gas
Applications
Life Science
Applications
Financial Services
Excel
Structural Analysis
Crash Simulation
Reservoir simulation
Seismic Processing
Structural Analysis
Crash Simulation
Portfolio analysis
Risk analysis
Compliance
Actual
Pricing
Modeling
Job Scheduler
App.exe
App.exe
Your applications
here
WCF Brokers
+
Resource allocation
Process Launching
Resource usage tracking
Integrated MPI execution
Integrated Security
App.exe
Interactive
Cluster
Applications
App.exe
WS Virtual Endpoint Reference
Request load balancing
Integrated Service activation
Service life time management
Integrated WCF Tracing
Service
(DLL)
Service
(DLL)
Service
(DLL)
Service
(DLL)
Service-Oriented Jobs
Job status
Service usage
report
Tracing logs
Admin
Experience
Node heat maps, Performance
monitor, & Event logs
Compute Nodes
Redundant Head Nodes
Control
Path
Job status
1
Create a
Session
Job Scheduler
Service
Router usage
Service
report
Job
Job
Tracing logs
User
Experience
2
Client
Data
path
3
Send/Receive
Messages
Start Broker
n Service Instances
Node
Manage
r
Node
Manage
r
Service
Instance
Service
Instance
…
Node
Manage
r
Service
Instance
WCF
Broker
WCF Broker Nodes
Occurring
on the
back end
Balance the requests
Grow & shrink service pool
Provide WS Interoperability
Track service resource usage
Run service as the user
Restart upon failure
Support application tracing
Job Templates
What are they?
• Encapsulates application resource requirements, job
prioritization and resource allocation policies
• Created by the admin, ACL’ed to a set of users so that the
resources can be delivered to the users in ways that meet the
organization’s business priorities
How to Define?
• Define the Value range: defines the range of values that the
term can take. The value range can be empty. If the value
provided at the job submission is not in the defined range, the
job is rejected.
• Provide a default value: if a profile defines the default value,
the user is not required to specify the job term
Job Templates
Admission control
1
Admin creates
resource
partitions by creating
job submission
policies
Descriptions
Runtime to be mandatory
A supercomputing center wanting to
enforce the runtime for all the jobs
Multiple Line of Businesses
(LOBs) sharing a cluster
Admin would like to apportion resources
to different nodes
Example
Policies
Power user job priority
Power user userA can use all the nodes in
the cluster
Definitions
Profile: default
Runtime:required
Default: none
Users: All
Profile LOB1:
Users: user1, user2
Priority: normal,
Select:”sas && ib && processorspeed > 2000000”
Uniform: switchId
Range: N/A
Profile LOB2:
Users: user3, user4
Askednodes:host2 host3 host 4
Profile PowerUser:
Users: userA
Askednodes: All
Priority: Highest
Range:
Job Submission
Permission
to Submit
Job Template
Cluster Resources
Job Submission
2
Users submit
using different
templates
Permission
to Submit
Job Template
Cluster Resources
Job Submission
Permission
to Submit
Job Template
Cluster Resources
Five New Scheduling Policies
• Resource Matching
– Job submit /nodegroup:appX myapp.exe
• Job Admission Policies via Templates
– Job submit /template:groupY myapp.exe
• Multi-Level Processor Allocation
– job submit /numsockets:4-8 myapp.exe
• Adaptive Allocation (Grow/Shrink)
– job submit myapp.exe
• Preemption
– Cluscfg PreemptionEnabled=true
Scenario: Placement via Job Context
node grouping, job templates, filters
Application
Aware
MATLAB
An ISV application (requires
Nodes where the application is
installed)
MATL
AB
10 GigE
InfiniBand
GigE
MMMMMM
A A A A A A
T T T T T T
L L L L L L
A A A A A A
B B B B B B
Blade Chassis
MMMMMM
A A A A A A
T T T T T T
L L L L L L
A A A A A A
B B B B B B
Capacity
Aware
Multi-threaded application
(requires machine with many
Cores)
8-core
servers
16-core
servers
32-core
servers
10 GigE
A big model (requires
Large memory machines)
InfiniBand
InfiniBand
GigE
C1
C0
C3
C2
M
P0
P1
Numa
Aware
M
||||||||
||||||||
M
M
4-way Structural Analysis MPI Job
M
M
M
M
||||||||
M
C0
C1
C2
C3
||||||||
P2
P3
IO
Quad-core
M
IO
32-core
M
Scenario : Memory Bus Saturation
Policy: Multi-level Allocation
Node 1
Node 2
P0
J1
J1
P1
P3
P2
J1
P1
P0
P0
J2
P3
J3
P1
P3
P2
J1: /numsockets:3 /exclusive: false
J3: /numsockets:3 /exclusive: false
S0
J2
P1
P2
S2
J2
S1
J2
P3
J2
J2
P1
P0
S3
J2
P3
J2
P1
P2
J2
P0
J2
P0
J2
P3
J2
P3
J2
P1
P2
S3
S2
P2
J2
P0
S1
S0
P2
P1
P0
P2
P3
J2: /numsockets:14 /exclusive: false
Scenario: mixed workload
Policy: Priority Resource Allocation
Faster turnaround for high priority jobs
Uses Grow/Shrink policy, Preemptive Scheduling Policies
Job 3 gets
Submitted
Processors
Job 3 gets
Completes
Job 1 Tasks
Job 2 Tasks
Job 3 High Priority
Time
Scenario: Long Tail Parametric Apps
Policy: Grow/Shrink Resource Allocation
Addresses “Long Tail” problem for jobs with many tasks of varying
runtime
Improve cluster utilization by giving nodes where they are needed
and to the highest priority jobs
Processors
Idle Processors
Idle Processors
Idle Processors
Job 1 Tasks
Job 2 Tasks
Job 3 Tasks
Time
Job Policies Summary
Features
Description
Version
Priority FCFS
Jobs are placed in the queue based on priority & submission Time. Jobs are in
the queue on a First-Come First-Serve basis in the order they are submitted,
except that ALL Highest Priority jobs are ahead of all Above Normal Priority jobs
which are ahead of all Normal Priority jobs etc . . .
Window CCS 2003
Windows HPC 2008
Backfilling
Resources are reserved based on resource allocations. If the Job Scheduler
identifies open windows for resources within the reserved timelines, it can
backfill by selecting smaller jobs for execution within this window
Window CCS 2003
Windows HPC 2008
Exclusive
Scheduling
When a job is “Exclusive” no other job can run on a node with that job. When a
task is “Exclusive” no other task can run on a node with that task.
Window CCS 2003
Windows HPC 2008
Resource Matchmaking
Schedule against admin defined groups of nodes. We allow scheduling based
on a limited amount of hardware properties and ordering by different hardware
properties.
Windows HPC 2008
Multi-level
Compute
Resource
Allocation
The Job Scheduler optimally places memory-intensive jobs to avoid contention
of memory, delivering maximum and predictable application performance.
Windows HPC 2008
Preemption
Pre-emption occurs when high-priority jobs take resources away from lowerpriority jobs which are already running.
Windows HPC 2008
Grow & Shrink
Scheduling
A job may not have “uneven resource requirements” it may just need 100
resources to run 1000 tasks. With Grow Shrink, the job scheduler can give it 1
resource to get it started, and then additional resources as they become
available. Once it has 100 resources and less than 100 tasks, job scheduler can
begin taking away the un-needed resources.
Windows HPC 2008
Interoperability & Open Grid Forum
What is it?
What is its value?
What’s the Status?
•A draft OGSA (Open Grid Services
Architectures) interoperability standard
for batch job scheduler task submission
and management
•Based on web services standards (HTTP,
XML, SOAP)
•Enables integration of HPC applications
executing on different platforms and
schedulers via web services standards
•Passed the public comment period
•Working on new extensions
LSF / PBS / SGE / Condor
Linux, AIX, Solaris
HPUX, Windows
Windows Cluster
Windows Center
Window Center
Job Scheduling Summary
Tools
End User Tools
Admin Tools
Developer
Tools
Scheduling
Policies
Windows CCS 2003
Windows HPC 2008
1. CLI
2. Job Manager UI
1. Powershell CLI
2. New Job Manager UI with built-in
parametric support & custom job
filtering
1. Cluscfg for configuration
2. clusrun
1. Configuration UI
2. Enhanced clusrun
1. C# API
2. COM API
1. WCF Integration
2. Scalable API with Rowset and eventing
support
3. Standard Job Submission Interface (HPC
Profile)
1.
2.
3.
4.
1.
2.
3.
4.
5.
FCFS
Backfill
Exclusive Scheduling
License scheduling
Resource matchmaking
Job Template
Multi-level processor allocation
Adaptive allocation / grow & shrink
Preemption
Parallel Programming
Parallel Programming
Parallel Program Tools
• Intel C++
• Intel Fortran
• PGI C++
• PGI Fortran
•
Compilers and
Languages
• Visual C++
• Visual C#
•Visual Basic
•Visual F#
Debuggers
• WinDbg
•VS Debugger (MC & MPI)
•Allinea Visual Studio plug-in (MPI)
•MPI/Event Tracing for Windows
•PGI MPI Debugger
Profilers
• Visual Studio
Profiler
• Vtune
• Code Analyst
•MPI/Event Tracing for
Windows
• PGI MPI Profiler
Analyzers
• Marmot
• MPI/Event
Tracing for
Windows
• Vampir
• Intel Trace
Collector/Analyzer
• Intel Thread Checker
• Utah U MPI model
checker
Parallel
Programming
Models
• OpenMP
•MPI (MS, Intel,
HP MPI Libs)
•MPI.NET
•MPI.C++
• PFx: Tark Paralell Library
• PFx: Parallel LINQ
• SOA on Cluster
•Intel Thread Building
Blocks
Math Libraries
• Intel MKL
• AMD IMSL
•Visual Numerics
• NAG
• Other OSS mathlibs
Available Now
–
–
•
Development and Parallel debugging in Visual
Studio
3rd party Compilers, Debuggers, Runtimes etc..
available
Emerging Technologies – Parallel Extensions
to .NET Framework
–
–
–
LINQ/PLINQ – natural OO language for SQL queries
in .NET
Task Parallel libraries
currently CTP June ‘08
Version Comparison
Feature
Windows Compute Cluster Server 2003
Windows HPC Server 2008
Windows Server 2003 SP1
Windows Server 2008 HPC Edition, Standard,
Enterprise, Datacenter
X64 (AMD64 or Intel EM64T)
X64 (AMD64 or Intel EM64T)
Memory
32 GB (Compute Cluster Edition)
128 GB (HPC Edition)
Node Deployment
Remote Installation Services(RIS)
Windows Deployment Services
N/A
Windows Failover Clustering and SQL Server Failover
Clustering
Basic node and job management
Integrated node and job management, grouping,
monitoring at-a-glance, diagnostics
Network Configuration Wizard
Improved Network Configuration Wizard
Winsock Direct-based
Network Direct-based. New shared memory
implementation for multicore processors
Operating system
Processor Type
Head Node Availability
Management
Network Topology
MS-MPI
Scheduler
Programmability
Reporting
Command line or GUI
Support for Batch or MPI based jobs
Integrated in management console, with full support
for Windows PowerShell scripting and legacy
command-line UI scripts from v1. Greatly improved
speed and scalability
Added support for interactive Service Oriented
Applications (SOA) using the Windows
Communication Foundation (WCF)
N/A
Integrated into Management console
Monitoring
Rely on Windows. No cluster specific support.
Heat map on cluster or node group. Per node charts.
Cluster-wide performance overview
Diagnostics
N/A
In the box verification tests and performance tests.
Store, filter, and view test results and history
Aggregate (Mb/s/core)
HPC Storage Solutions
Shared File
Systems or SAN
file systems
Parallel File
Systems
• IBM – GPFS
• Panasas – Active Scale
• SUN - Lustre
• HP - PolyServe
• Ibrix - Fusion
• Quantum - StorNext
• SANbolic – Melio file system
NAS and
Clustered
NAS
• Windows Server 2003
• Windows Server 2008
…
Number of cores in cluster
Greater
Sophistication
Release Schedule
Beta 1
Nov
2007
CTP 1
March
2008
CTP2
April
2008
• RC1– Publicly Available
• RTM – Early Fall 2008
Beta 2
May
2008
RTM
Fall
2008
Industry Focused Partners
Windows HPC Server 2008 Offerings
Basic Turnkey
•
•
Turnkey Solutions up to 1000
nodes
HPC Pack
•
•
•
•
High Availability
Management
Deployment
Job Scheduling
•
•
•
3rd Party packages for enhanced
job scheduling, policies
Enterprise Management
•
•
No single point of failure
Windows Server 2008
Enterprise for head node high
availability
SQL Server 2008 Enterprise for
high availability
•
Scale to many 1000’s of nodes
System Center Operations
Manager for enhanced
monitoring
System Center Configuration
Manager for enhanced
deployment and patch
management
Systems Center
Configurations Manager
Systems Center
Operations Manager
SQL Server 2008
Standard
SQL Server 2008 Express
3rd
Party
HPC Pack
SQL Server 2008
Standard
3rd
Party
HPC Pack
Windows HPC Server 2008
3rd
Party
Windows HPC
Server 2008
Windows Server 2008 Core
HPC Pack
Windows
Server 2008
Enterprise
Windows HPC
Server 2008
•
•
•
Windows
Server 2008
Enterprise
Lightweight small footprint/No UI/CLI
HyperV
No .NET Framework
Windows HPC Roadmap
Mainstream HPC
Mainstream High Performance Computing on Windows platform
 Interoperability: Web Services for Job Scheduler, Parallel File Systems
 Applications: Service Oriented, Batch, .NET
 Turnkey: Enabling pre-configured OEM solutions
 Scale: Large scale, non-uniform clusters, diagnostics framework
Service Pack 1
Web Releases
 Performance & Reliability Improvements
 Support for Windows Server 2003 SP2
 Support for Windows Deployment Services
 Vista Support for CCP Client tools
 MOM Pack
 PowerShell for CLI
 Tools for Accelerating Excel
Mainstream High Performance Computing on Windows platform
 Simple to set up and manage in familiar environment
 Integrated with existing Windows infrastructure
Version 2
H2 2008
SP1 & Web
2007
V1
Summer 2006
Resources
• Microsoft HPC Web site – Evaluate Today!
– http://www.microsoft.com/hpc
• Windows HPC Community site
– http://www.windowshpc.net
• Windows HPC Techcenter
– http://technet.microsoft.com/en-us/hpc/default.aspx
• HPC on MSDN
– http://code.msdn.microsoft.com/hpc
• Windows Server Compare website
– http://www.microsoft.com/windowsserver/compare/default.mspx
© 2008 Microsoft Corporation. All rights reserved.
This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.
About the MPI Forum
•
•
•
•
•
The MPI forum is an international forum that defines the MPI standard. Since 1992 is has
defined MPI 1.0 1.2 and 2.0 standards. This standard body is important as over the years
MPI has become the de-facto parallel programming interface/paradigm.
The MPI forum has come together this year, 2008 in an effort to renew and update the
MPI standard to put MPI 2.1, 2.2 and 3.0 standards. The MPI forum is determined goal is
to answer the world demand for significant features which will enable MPI to better
handle the scope of tomorrow’s systems and the breath of applications using MPI.
The MPI forum body includes representatives from many software companies,
Universities, MPI implementers and hardware vendors. The forum meets in person every
two months to discuss proposals, issues at hand and vote on the outcome. Microsoft
takes a leading role in this forum leading several workgroup and hosting some of the
forum meetings.
The MPI forum goal is to release MPI 2.1 standard by the end of 2008, MPI 2.1 by mid
2009 and MPI 3.0 by the end of 2010.
The unofficial wiki page web site; you can find MPI workgroups information on these
pages.
–
•
The official MPI web site
–
•
https://svn.mpi-forum.org/trac/mpi-forum-web
http://www.mpi-forum.org
The meetings info web pages
–
http://meetings.mpi-forum.org
Windows HPC Success Stories: Financial Services
Financial Services Firm Scales Fast-Growing Hedging Application on Low-Cost
Windows Cluster
Fund Manager Improves Performance, Eases Administration with Computing Cluster
“It’s like having an incredibly powerful CPU at your finger tips. The ease of use creates a kind of
transparency where the line between the desktop and the compute cluster no longer exists.” Christopher Mellen, Head of Research, Grinham Managed Funds
Investment Bank Improves Competitive Edge with High Performance Computing
Investment Bank Plans to Boost IT Performance, Reduce Hardware Costs with New
Server Technology
“From now on, whatever we will be testing, developing, or evaluating will be based on Windows
Server 2008—whether it’s business-driven, transition-driven, or research-driven.” - Sorin Manta,
Manager, Windows Server Infrastructure, Technology and Operations, BMO Capital Markets
IT Services Company Develops Risk and Trading Systems on Integrated Platform
ISV Opts Against Open Source, Doubles Revenue with Office Business Application
“By using Microsoft technologies as the foundation for our development, we’re producing an easyto-use, understandable, low-cost offering that can be confidently adopted by financial services
customers.” - Graham Twaddle, Founder and Chief Executive Officer of Corporate Modelling
Windows HPC Success Stories: Manufacturing
Switch from Linux to Windows Increases Value of HPC for Golf Equipment
“Manufacturer jobs that took 40 hours are now down to 8, enabling engineers to test and refine
their designs much faster.” - John Loo, Design Systems Senior Manager, Callaway Golf
Leading HPC Vendor Eases Adoption for Customers, with New Windows-Based
Offerings
“Many companies see value from HPC, but they have little to no experience with Linux… We need
to provide a way for them to advance their theories and research using what they already know.” Beverly Bernard, Product Manager, SGI
64-bit Compute Cluster Performs Crash Analysis, Aides in Safety Compliance
Shipbuilder Introduces HPC on Microsoft to Develop High Value-Added Ship Types
Boeing Tests High Performance Computing Cluster, Improves Processing Time
“Because we develop intelligent graphics software in a Windows environment, it makes sense to
work with an HPC cluster that supports that framework.” - John F. Bremer Advanced Computing
Technologist, Intelligent Graphics Group Boeing
Manufacturer Provides Engineers with Easy-to-Use, High-Performance Computing
Solution
Engineers at French Production Company Improve Productivity with New Business
Solution
Windows HPC Success Stories: Manufacturing
Australian Company Delivers Solutions Faster, Expands Capabilities with HPC
Solution
“I came in with zero knowledge of Windows HPC Server 2008 deployment, although I knew a lot
about Linux. Within a couple of days, I was deploying Windows-based nodes.” - Dr. Simon Beard
,Systems Specialist, On Demand Group, ISA Technologies
NASCAR Team Turns to High Performance Computing to Sharpen Competitive
Edge
“With simulation times reduced from 24 hours to about 30 minutes, we can run multiple simulations
for each race and better tune the simulations for each car, track, and set of track conditions.” Mark Paxton
Research and Development Engineering Manager, NASCAR Team, Chip Ganassi Racing
Pebble Bed Modular Reactor (PBMR) in South Africa adopts Windows HPC
Technology Over Linux.
SIMULIA Delivers Simulation Solutions Faster with Windows HPC Server 2008, and
Sees the Advanced Programming Tool Set as a Critical Asset.
Microsoft High Performance Computing Solution Helps Oil Company Increase the
Productivity of Research Staff
“With Windows Compute Cluster Server, setup time has decreased from several hours—or even
days for large clusters—to just a few minutes, regardless of cluster size.” - IT Manager, Petrobras
CENPES Research Center
Windows HPC Success Stories: Manufacturing
French Yacht Team Streamlines Design with Secure and Familiar Technology
“The transition from Linux to Windows Compute Cluster Server 2003 was flawless. In fact, it was
so easy we didn’t even notice a change in the office.” - Bernard Nivelt, Lead Designer, AREVA
Challenge
Easier Cluster Management Helps Northrop Grumman Improve Productivity
“Windows Compute Cluster Server has caused a paradigm shift. Before I had to limit my problem
size because I ran out of resources. Now I feel enabled to think bigger.” - Thi Pham, PhD, Systems
Engineer, Space Technology Sector, Northrop Grumman
Simulation Software on x64 Compute Clusters Boosts Performance, Reduces Costs
Florida Boosts Productivity, Cuts Run Times with High-Performance Computing
Cluster
Software Company Delivers 64-Bit Fidelity and Speed for Computer-Aided
Engineering
Windows HPC Success Stories: Science, Education,
Research
Windows HPC Server 2008 Ranks at 23 Among the World’s TOP500 Largest
Supercomputers with 68.5 TFlops and 77.7 Percent Efficiency. (June 2008)
“The performance of Windows HPC Server 2008 has yielded efficiencies that are among the
highest we’ve seen for this class of machine.” - Robert Pennington, Deputy Director, National
Center for Supercomputing Applications
The Umeå Cluster Achieved a LINPACK Score of 46.04 TFlops and 85.6 Percent
Efficiency, Making Their System the Second Largest Windows Cluster Ever
Deployed and the Fastest Academic Cluster in Sweden. (June 2008)
Leading Supercomputing Center in Italy Eases Use, Improves Access with New
Cluster
“It will be a big benefit for us to offer researchers a high-performance computing resource with a
familiar interface and a natural, user-friendly way to use the cluster from home.” - Dr. Marco Voli,
CINECA
Researchers’ Move from Linux to Windows Yields Performance Gains, New
Capabilities
“We were quite surprised when, without any optimization, the new Windows–based HPC system
outperformed our highly optimized Linux cluster.” – Valerie Daggett, Professor, University of
Washington
Facility for Breakthrough Science Seeks to Expand User Base with Windows-Based
HPC
Windows HPC Success Stories: Science, Education,
Research
Early Detection of Cancer One Step Closer to Solution with Microsoft, Dell and Intel
“The user interface and structure of the Microsoft Windows Compute Cluster Server make
managing a large, high-performance computing cluster far less daunting than with other operating
systems.” - Dr Robert Moritz, Manager of the Proteomics Facility at LICR and Director of the
Australian Proteomics Computational Facility
National IT Center Improves Customer Service with High-Performance Compute
Cluster.
“Using Windows for high-performance computing means we can offer our customers real added
value.” - Uwe Wössner, Head of the Visualization Department, High Performance Computing
Center Stuttgart
Inventor of Beowulf Cluster Exposes Young Minds to High-Performance Computing
Portuguese University Accelerates Cancer Research with High-Performance
Computing
Microsoft Researchers Boost Task Productivity Fiftyfold with Cluster Server Software
Windows HPC Success Stories: Science, Education,
Research
Supercomputing Solution Reduces IT Administration Needs at University of
Cincinnati Genome Research Institute.
Research-Driven University Breaks Down Barriers to High-Performance Computing
Environmental Scientists Join Forces Against Climate Change with Integrated
Platform
“This is the first time we’ve delivered an integrated solution whereby researchers can sit in front of
a Web browser and drive it to completely different scenarios using the data and models of different
institutions.” - Simon Cox, Professor of Computational Methods at the University of Southampton
and Technical Director for Genie
Scientists Accelerate Research and Insight with Accessible, High-Performance
Computing Environment
“Even students who come from a Linux background and are using Microsoft developer tools for the
first time are finding the change to be positive.” - Iain Buchan, Director of the Northwest Institute for
BioHealth Informatics at the University of Manchester
University of British Columbia (Vancouver) selects Windows HPC technology over
Linux for Masters of Digital Media graduate program
Windows HPC Success Stories: Digital Content
Creation
Digital Marketing Firm Installs Microsoft HPC Solution to Simplify IT Operations
Digital Media School Deploys Render Farm Technology, Cuts Compute Runtime By
Days
“Before we had render farm, every student rendered on his or her own PC, so sharing images and
viewing the current status was not easy. Now we can decide how many PCs will render a
particular image. On 32 machines it takes just a couple of hours—this is a huge reduction in time.”
- Ng Kian Bee, Deputy Director, Games & Digital Entertainment of NYP’s SIDM