Trish Damkroger Vice President, Data Center Group General … · 2017-09-01 · High Performance...
Transcript of Trish Damkroger Vice President, Data Center Group General … · 2017-09-01 · High Performance...
Trish DamkrogerVice President, Data Center GroupGeneral Manager, Technical Computing Initiative
Before We Begin - The coming flood of dataBy 2020…
The average internet user will generate
~1.5 GB of traffic per daySmart hospitals will be generating over
3,000 GB per day
A connected plane will be generating over
40,000 GB per dayA connected factory will be generating over
1,000,000 GB per day
Self driving cars will be generating over
4,000 GB Per day … Each
The world is changing – without Analytics and aI workloads we will drown in data!!!
Different Systems (Today)
FORTRAN / C++ ApplicationsMPI
High Performance
Java* ApplicationsHadoop*
Simple to Use
SLURMSupports large scale startup
YARN*More resilient of hardware failures
Lustre*Remote Storage
HDFS*, SPARK*Local Storage
Compute & Memory FocusedHigh Performance Components
Storage FocusedStandard Server Components
Server StorageSSDs
SwitchFabric
Infrastructure
ProgrammingModel
Resource Manager
File System
Hardware
Server StorageHDDs
SwitchEthernet
Infrastructure
3Daniel Reed and Jack Dongarra, Exascale Computing and Big Data in Communications of the ACM journal, July 2015 (Vol 58, No.7), and Intel analysisOther brands and names are the property of their respective owners.
Emerging Real-Time Workflows
Small Data + Small Compute
e.g. EDA
Big Data + Small Compute e.g. Search, Streaming, Data Preconditioning
Small Data +Big Compute
e.g. Mechanical Design, Multi-physics
Da
ta
Compute
4
One System Architecture?
From Here … …TO?
5
The Challenges of Moving to (HPC && Big Data)
Source: NERSC Workload Analysis on Hopper. K. Antypas et al. 2014
Many Codes on a System
Over 600 Different Applications on NERSC’s Hopper System
6
The Challenges of Moving to (HPC && Big Data)
Source: NERSC Workload Analysis on Hopper. K. Antypas et al. 2014
Many Codes on a System
Over 600 Different Applications on NERSC’s Hopper System
Source: Intel for illustration purposes
Varied Resource Needs
Typical HPC Workloads
High Frequency
Trading
Numeric Weather
Simulation
Oil & Gas Seismic
Sy
ste
m c
ost
ba
lan
ce
Typical Big Data Workloads
Video Survey Traffic Monitor
Personal Digital Health
Sy
ste
m c
ost
ba
lan
ce
Processor Memory Interconnect Storage
7
Benefits of a Single System Architecture
Flexibility to manage workflows and maximize system utilization
234
1
Improved Total Cost of Ownership
Overcomes cost and complexity of moving data
Potential for a unified SW ecosystem
8
Realizing the Benefits Requires Coordinated Design
ComputeMany core
Multicore
Intel Graphics
Intel Xeon + FPGA
System Software StackSimplifies Use and Application Development
InterconnectConfigurable
Next-Generation Fabric
Memory/StorageConfigurable High BW Memory
Non-Volatile MemoryStorage Burst BufferLustre File System
9
Converged Architecture for HPC and Big Data
10
FORTRAN / C++ ApplicationsMPI
High Performance
Java* ApplicationsHadoop*
Simple to Use
Lustre* with Hadoop* AdapterRemote Storage
Compute & Big Data CapableScalable Performance Components
Server Storage(SSDs and Burst Buffers)
Intel®Omni-Path
Architecture
Infrastructure
Programming Model
Resource Manager
File System
Hardware
*Other names and brands may be claimed as the property of others
HPC & Big Data-Aware Resource Manager
Key Enabler: A Configurable Memory-Storage Hierarchy
11
Processor
Compute Node
I/O Node
Remote Storage
Compute
Today
Caches
Local Memory
SSD Storage
Parallel File System(Hard Drive Storage)
Hig
he
r B
an
dw
idth
. L
ow
er
La
ten
cy a
nd
Ca
pa
city
Some remote data moves onto I/O node
I/O Node storage moves to compute node
Local memory is now faster & in processor package
Compute
Future
Caches
Non-Volatile Memory
Burst Buffer Storage
Parallel File System (Hard Drive Storage)
In-Package High Bandwidth Memory*
*cache, memory or hybrid mode
MachineLearning
Modeling & Simulation
High PerformanceData Analytics
Many workloads | One architectural framework
Small Clusters Through Supercomputers
Standards-Based Programmability
On-Premise and Cloud-Based
Intel® Scalable System Framework
Compute and Data-Centric
Legal Disclaimers
Intel technologies features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on
system configuration. No computer system can be absolutely secure. Check with your system manufacturer or retailer or learn more at [intel.com].
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark,
are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should
consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with
other products.
All information provided here is subject to change without notice. Contact your Intel representative to obtain the latest Intel product specifications and roadmaps.
Results have been estimated or simulated using internal Intel analysis or architecture simulation or modeling, and provided to you for informational purposes. Any differences in
your system hardware, software or configuration may affect your actual performance.
Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on
system configuration. No computer system can be absolutely secure. Check with your system manufacturer or retailer or learn more at https://www-
ssl.intel.com/content/www/us/en/high-performance-computing/path-to-aurora.html.
Intel, the Intel logo, Xeon, and Intel Xeon Phi are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States or other countries.
*Other names and brands may be claimed as the property of others.
© 2015 Intel Corporation
13