SUSE® High Performance Computing

Kai Dupke Meike Chabowski Senior Product Manager Senior Product Marketing Manager SUSE Enterprise SUSE Linux Enterprise [email protected] [email protected]

Distribution: pdf any Date: 2015-10-22 Not a public document. 3 HPC Impacts Our Lives – Space Research

4 5 HPC Impacts Our Lives – Weather & Climate

6 7 HPC Impacts Our Lives – Oil & Gas

8 9 HPC Impacts Our Lifes - Entertainment

10 From Cinema ...

… to Games

11 High Productivity Computing Big Data – or HPC??

12 Overview HPC Overview

SUSE® High Performance Computing

• Solving computational, data-intensive, or numerically-intensive tasks • Reducing the time and effort required to set-up and maintain HPC clusters • Ensuring that all components of the HPC stack work together

14 HPC Development

SUSE® High Performance Computing

• Yesterday

‒ Academia and Research • Today

‒ Academia and Research ‒ Financial Services ‒ Oil and Gas ‒ Semiconductor ‒ Life Sciences ‒ Manufacturing • Tomorrow

‒ Departmental and workgroup clusters ‒ High Productivity Computing

15 Market Segmentation SUSE High Performance Computing

System HPC class Ready Key drivers GTM Budget Super Computer >500K$ +++ Special build HW HPC-IHV 'top 500' only performance count Self-supported Partner supported Divisional <500K$ +++ Customized HW IHV Partner driven ISV Partner supported SI SUSE supported Departmental <250K$ ++ Commodity HW Channel Business driven SUSE SUSE supported

Work Group <100K$ + Customer driven Channel Home brewed Shop

16 Split Market

SUSE® HPC

Commercial Scientific High Productivity Top 500-class Computing

• Lighthouse projects • Highly specialized application • Government sponsored • ROI and reliability • Generic workloads are key • Often self-supported by • Data Center support Academic staff • Commodity hardware • Specialized hardware

17 SUSE Linux Enterprise HPC SUSE – Strong in HPC Market!

SUSE® HPC

MULTI- and MANY-CORE PROCESSOR SUPPORT , AMD, POWER ….. COOPERATION TECHNOLOGY IBM Bull Kernel 3.x SUSE SGI NEC Lustre enablement Since 1992 HP Ceph storage platform Cray up to 8192 cores Strong Dell Cisco ….. in Top500 ….. …..

ACADEMIC HIGH PRODUCTIVITY AND RESEARCH COMPUTING LRZ / SuperMUC Total BSC / MareNostrum Baker Hughes Tokyo Institute of Technology Texas Instruments Beijing Computing Center ….. NASA …..

19 Why Linux?

SUSE® High Performance Computing

• Open Source benefits

‒ Easy to customize, maintain and improve • Innovation

‒ Beowulf Clusters “born” on Linux • Modularity

‒ GUI overhead not required

‒ appliance form factors • Linux Standards

‒ Large base of tools, including remote management

‒ Hardware availability

‒ Large vendor ecosystem surrounding Linux HPC clusters

20 Linux Preferred for HPC

SUSE® High Performance Computing

• Linux

‒ runs on more than 97% of the world's top 500 supercomputers*

‒ is used by nearly 90% of general clusters

‒ Linux is used in the majority of HPC systems, from smaller departmental implementations to larger, integrated cluster solutions

*top500.org July 2015

21 Why SUSE® Linux Enterprise Server For High Performance Computing

• Early player in HPC, pushing innovation and new technologies • Highly reliable, interoperable and manageable server operating system • Built to power mission-critical workloads in physical, virtual and cloud environments

• The natural successor to UNIX, backed by proven services for UNIX migration • Special features to improve performance • Backed by established ecosystem – support and certificates • The only Linux recommended by Microsoft

22 SUSE Additional Features

SUSE® High Performance Computing

• Up-to-date Linux Kernel for optimal performance • CPU Management and System Activity

‒ CPUset System, CPUset command line tool

‒ Sysstat package

‒ IRQbalance • OpenFabrics Enterprise Distribution (OFED)

‒ Remote Direct Memory Access (RDMA) switched fabric technologies, high-speed data transport technologies for server and storage connectivity • SystemTap, LTTng 2.0 • Lustre enabled Kernel

23 SUSE Advanced I/O Processing

SUSE® High Performance Computing

• Asynchronous I/O (AIO)

‒ Input/output processing that permits other processing to continue before the transmission has finished • Modular I/O Scheduler

‒ Algorithm most suitable for workload can be chosen dynamically • Multi-core/hyper-threading processor support

‒ Execute threads in parallel within each individual processor

‒ Supports up to 4096 cores per system • Intel I/O Acceleration

‒ Offloads the CPU towards the network card, thus allowing the system to continue processing data while I/O is taking place

24 Update Simplify projects!

SUSE® HPC

• Simplified model

‒ Number of socket pairs matter

‒ Socket pairs are accumulated per system

‒ Head nodes and compute nodes are threaten equal

26 Example – regular HPC setup

SUSE® HPC

• 2 (8 sockets) head nodes for redundancy / scalability 100 (4 sockets) compute nodes • 416 sockets total (order: 208x 1-2 sockets)

Head Node Compute Nodes total Client 2x 8 sockets + 100x 4 sockets = 416 sockets

27 Keep it running!

SUSE® HPC

• SUSE Vendor Support

‒ Maintenance

‒ Standard & Priority support for the whole system

29 Recent Developments

SUSE® High Performance Computing

• Storage • SUSE Linux Enterprise

‒ Release of SUSE Storage ‒ 11 SP4 with latest HW with Infiniband support enablement (Intel, Power8) • ARM64 • Network

‒ Partnering with Cavium ‒ Higher network throughput

‒ SUSE Linux Enterprise for ‒ Added tunables in the IP ARM64 stack (for lower latency) • Cloud

‒ MS Azure with SLES 12: RDMA & Infiniband

30 Partner, Customers Customers and Partners

SUSE® High Performance Computing Customers

Partners

32 PANGEA – Total Exploration SUSE High Performance Computing

• Oil & Gas exploration

‒ Process seismic data

‒ Simulation of deposit fluids

• Superior Performance “... from our point of view, SGI plus SUSE Linux ‒ 2.3 petaflops Enterprise Server was a complete, integrated ‒ 10x performance increase solution.”

‒ Equivalent to 27,000 PCs “... SUSE Linux Enterprise Server gives us the ability to • Future is coming keep scaling on ever larger machines.” ‒ 6.7 petaflops — Diego Klahr HPC Engineer Total ‒ Equivalent to 80,000 PCs

33 Intel Cluster Ready Program

SUSE® High Performance Computing

• Designed to simplify purchasing, deployment and management of HPC clusters • SUSE Linux Enterprise Server is Intel Cluster Ready and powers many certified Intel Cluster Ready systems • intel® Cluster Ready “recipes” are available with SUSE Linux Enterprise Server

‒ Reference designs to help hardware vendors, platform integrators, and system integrators design and build certified Intel Cluster Ready systems

36 Outlook Challenge

SUSE® High Performance Computing

HPC market fast developing

Stack components provided by various vendors

Some stack components run in parallel

Mix of small and big vendors

Segmented into commercial and scientific

38 Outlook

SUSE® High Performance Computing

• SUSE Linux Enterprise

‒ 12 SP1 beta program running • SUSE HPC

‒ Evaluating optimized SUSE Linux Enterprise for HPC

‒ Your input is needed!

39 Forward looking statement, might change without notice. Learn more

www.suse.com/products/server/hpc.html

Thank you.

Kai Dupke Meike Chabowski Senior Product Manager Senior Product Marketing Manager SUSE Linux Enterprise SUSE Linux Enterprise [email protected] [email protected]

40 Backup HPC Stack HPC Stack

SUSE® High Performance Computing

Application

Queuing / Management Software & Tools PBS Pro Moab IBM LSF Bright CM

Message Passing Interface Storage Network MPI EXT3 XFS BTRFS 10G OFED Parastation Intel HP SGI OCFS2 NFS pNFS TCP offload MPICH openMPI IBRIX GPFS pNFS Lustre cephFS

SUSE Linux Enterprise Server

Hardware

= SUSE supported = SUSE Partner = SUSE future

43 SuperMUC SuperMUC – Facts

SUSE® High Performance Computing

• 60x faster, one of the fastest HPC systems in Europe • 20x better performance per Watt, provide green HPC • > 155,000 Intel Processor, migration from Itanium2 to

45 LRZ - Leibniz Rechenzentrum Europe’s supercomputer run SUSE Linux Enterprise Server

Business challenge: LRZ is part of the Gauss Centre for Supercomputing (GCS), which “We have relied on SUSE operates the most powerful HPC infrastructure in Europe, and needs to Linux Enterprise Server for 15 provide researchers across Europe with a reliable and powerful HPC platform, which enables users to make faster progress in their complex years, and have always been research projects. To reduce the environmental impact of HPC, the institution very satisfied. aimed at improving the energy efficiency leverage established automation solutions to maximise the efficiency and manageability of the new The SUSE team is close at supercomputing platform. hand, should we require support or guidance. Solution: Working with SUSE and IBM, LRZ implemented SuperMUC with approx. 9,400 general purpose computing nodes, a peak performance of three We have received highly Petaflop/s, comprised of 155,000 Intel Xeon processor cores and more competent support over the than 300 TB main memory. LRZ chose to run SuperMUC on SUSE Linux years, and look forward to Enterprise Server, leveraging SUSE’s proven HPC expertise and leading collaborating with them. automation tools such as AutoYaST, which allows systems to be installed without manual intervention. — Dr. Herbert Huber Division Head of Supercomputing Benefits: Leibniz Rechenzentrum • Completed easy and smooth migration from previous 2 infrastructure to new x86 processor architecture • Considerably simplified configuration and automation of the new system, using the automation capabilities of AutoYaST(integrated with SLES) • Improved the energy efficiency: SuperMUC delivers appro. 20 times more performance per watt than its predecessor • Boosted overall performance by a factor of 60

46 SuperMUC – Facts and Business Aspects

SUSE® High Performance Computing

47 SuperMUC – System Overview

48 SuperMUC – System Overview

49 SuperMUC – Business Aspects SUSE High Performance Computing

• Hot Water Cooling – reduce cooling cost

‒ Use free air cooling

‒ Use of system heat for heating and technical processes • RAS driven – high system availability

‒ Full maintained SUSE Linux Enterprise Server

‒ Full support via IBM and SUSE • Automated deployment – less management cost

‒ Full use of SUSE's autoYAST feature

50 SuperMUC – SUSE benefits SUSE High Performance Computing

• Support for Itanium2 and x86 “We have relied on SUSE Linux Enterprise Server for ‒ Smooth migration of old to new system 15 years, and have always been very satisfied. ‒ No additional staff training needed The SUSE team is close at • Great support experience hand, should we require support or guidance. ‒ Cooperation for more than 15 years We have received highly ‒ Backed by SUSE's winning support competent support over the years, and look forward to • collaborating with them. Easy deployment methods — Dr. Herbert Huber ‒ SUSE's autoYAST used today Division Head of Supercomputing ‒ Other SUSE offerings – SUSE Cloud, SUSE Leibniz Rechenzentrum Manager – considered

51 Corporate Headquarters +49 911 740 53 0 (Worldwide) Join us on: Maxfeldstrasse 5 www.suse.com www.opensuse.org 90409 Nuremberg Germany

52 Unpublished Work of SUSE. All Rights Reserved. This work is an unpublished work and contains confidential, proprietary and trade secret information of SUSE. Access to this work is restricted to SUSE employees who have a need to know to perform tasks within the scope of their assignments. No part of this work may be practiced, performed, copied, distributed, revised, modified, translated, abridged, condensed, expanded, collected, or adapted without the prior written consent of SUSE. Any use or exploitation of this work without authorization could subject the perpetrator to criminal and civil liability.

General Disclaimer This document is not to be construed as a promise by any participating company to develop, deliver, or market a product. It is not a commitment to deliver any material, code, or functionality, and should not be relied upon in making purchasing decisions. SUSE makes no representations or warranties with respect to the contents of this document, and specifically disclaims any express or implied warranties of merchantability or fitness for any particular purpose. The development, release, and timing of features or functionality described for SUSE products remains at the sole discretion of SUSE. Further, SUSE reserves the right to revise this document and to make changes to its content, at any time, without obligation to notify any person or entity of such revisions or changes. All SUSE marks referenced in this presentation are trademarks or registered trademarks of Novell, Inc. in the United States and other countries. All third-party trademarks are the property of their respective owners.