Ce diaporama a bien été signalé.
Nous utilisons votre profil LinkedIn et vos données d’activité pour vous proposer des publicités personnalisées et pertinentes. Vous pouvez changer vos préférences de publicités à tout moment.
Deploying massive scale for
graphs for realtime insights
B Brech
CTO POWER Solutions
blbrech@us.ibm.com
2
2
© 2016 International Business Machines Corporation
Drive Efficiency
- Time Reduction
- Cost Reduction
- Consistency
Be...
3
3
© 2016 International Business Machines Corporation
1990’
s
2020’
s
Video
Text
Exa
Pet
a
Ter
a
Gig
a
DataVolume
2000’
s...
4
4
© 2016 International Business Machines Corporation
Time is Money
and
Insights are critical
Ingest Analyze Act Measure ...
5
5
© 2016 International Business Machines Corporation
Recommendation engines
- used in variety of industries
Network intr...
6
6
© 2016 International Business Machines Corporation
DB2 > DB2Blu
SAP > SAP Hana
Oracle > 12C
CICS
EnterpriseDB
Etc..
No...
7
7
© 2016 International Business Machines Corporation
Built with open innovation to
put your data to work across the ente...
8
8
© 2016 International Business Machines Corporation
UNSTRUCTURED IN-MEMORY STRUCTURED
Flash for extreme
performance
Mas...
9
9
© 2016 International Business Machines Corporation
POWER Ecosystem
Designed
for Big Data
Workload
Acceleration
Defined...
10
10
© 2016 International Business Machines Corporation
Fundamental forces are accelerating
industry change
IT innovation...
11
11
© 2016 International Business Machines Corporation
NVLINK
GPUFPGA
Flash NIC
MRAM PCM
Solution Acceleration is a key ...
12
12
© 2016 International Business Machines Corporation
NVLINK
GPU
Flash
Graphics – CAE - EDA
Weather
Defense
Financial S...
13
13
© 2016 International Business Machines Corporation
IBM Data Engine for NoSQL is an integrated platform for large and...
14
14
© 2016 International Business Machines Corporation
Identical hardware with 3 different
paths to data
FlashSystem
Con...
15
15
© 2016 International Business Machines Corporation
ON
Efficient IO Enables True Utilization
of Storage Bandwidth
 U...
16
16
© 2016 International Business Machines Corporation
Neo4j + IBM POWER8:
Unparalleled Scale and Performance
Neo4j on I...
© 2016 IBM Corporation
Real-World mixed graph transaction workload
running Neo4j on POWER8 delivers 1.82X better
performan...
© 2016 International Business Machines Corporation 18
Scale up and/or out based on your
application requirements
• Out-of-...
Open innovation to put data to work
across the enterprise
Thanks!
© 2016 International Business Machines Corporation
19
© Copyright International Business Machines Corporation 2016
Printed in the United States of America September 2016
IBM, t...
Prochain SlideShare
Chargement dans…5
×

Deploying Massive Scale Graphs for Realtime Insights

224 vues

Publié le

Graph databases have been at the forefront of helping organizations manage and generate insights from data relationships, and applying those insights in real-time to drive competitive advantage. As organizations gain value in deploying graph databases, the data volumes managed are growing exponentially pushing the limits of large-scale in-memory graph processing. Neo4j and IBM Power Systems combined forces to deliver a market leading scalable graph database platform capable of affordably storing and processing graphs of extremely large size and offering real-time insights, using flash and FPGA accelerators. In this session we will cover the use cases driving the need for this extremely scalable platform and how this platform offers an easy to deploy model for extreme scale graph databases.

Publié dans : Logiciels
  • Soyez le premier à commenter

  • Soyez le premier à aimer ceci

Deploying Massive Scale Graphs for Realtime Insights

  1. 1. Deploying massive scale for graphs for realtime insights B Brech CTO POWER Solutions blbrech@us.ibm.com
  2. 2. 2 2 © 2016 International Business Machines Corporation Drive Efficiency - Time Reduction - Cost Reduction - Consistency Better Insights - Broader Scope - Learning Models - Speed & Accuracy Better Business - Innovation - Customer Care - Reactivity Business relies more on Data than ever before
  3. 3. 3 3 © 2016 International Business Machines Corporation 1990’ s 2020’ s Video Text Exa Pet a Ter a Gig a DataVolume 2000’ s 2010’ s Structured data Audio Image Me d High Lo w ComputationalNeeds SophisticationofAnalysis Expressiveness Digital Marketing 10+% of video views Wide Area Imagery 100’s TB per day72 video hrs/minute Media Source: IBM Market Insights based on composite sources Safety / Security Healthcare Customer 1B camera phones 1B medical images/yr 10s millions cameras Enterprise Video Used by 1/3 of enterprises Data Volume Data Velocity Data Authenticity Data Complexity Data Variability Data Variety While Data is Exploding
  4. 4. 4 4 © 2016 International Business Machines Corporation Time is Money and Insights are critical Ingest Analyze Act Measure Learn Optimize Decision time is shrinking
  5. 5. 5 5 © 2016 International Business Machines Corporation Recommendation engines - used in variety of industries Network intrusion prevention Fraud prevention Financial Services BioMedical - Genomics Combination of Scale & Speed is critical in many use cases Extreme Scale Example: - 30TB and growing DB - 25 BG/s ingress - over 400K updates / Sec - 60B+ relationships - Query Response < 200ms
  6. 6. 6 6 © 2016 International Business Machines Corporation DB2 > DB2Blu SAP > SAP Hana Oracle > 12C CICS EnterpriseDB Etc.. NoSQLs : MemCached, REDIS, NEO4J, CASSANDRA, MARIA, MONGO, ORIENT, COUCH, Etc… Traditional DBs going in-memory Designed as in-memory repositories AnalyzeDecision Innovation Act Ingest But in-memory has some constraints and limits. Data repositories are changing also
  7. 7. 7 7 © 2016 International Business Machines Corporation Built with open innovation to put your data to work across the enterprise Designed for Big Data Open Innovation Platform Superior Cloud Economics IBM POWER8 : Designed for Big Data
  8. 8. 8 8 © 2016 International Business Machines Corporation UNSTRUCTURED IN-MEMORY STRUCTURED Flash for extreme performance Massive IO bandwidth Continuous data load Parallel processing Large-scale memory processing Optimized for a broad range of big data & analytics workloads: Processors flexible, fast execution of analytics algorithms Memory large, fast workspace to maximize business insight Cache ensure continuous data load for fast responses 4X threads per core vs. x86 (up to 1536 threads per system) 4X memory bandwidth vs. x861 (up to 16TB of memory) 4X more cache vs. x862 (up to 800MB cache per socket) IBM POWER8 brings performance and scale
  9. 9. 9 9 © 2016 International Business Machines Corporation POWER Ecosystem Designed for Big Data Workload Acceleration Defined by Software Retail Healthcare Banking Government Telecom Open and Collaborative Technology & Price/Perf Leadership Watson LinuxHadoop POWER8 Hypervisor Virt I/O Server Shared I/O Single SMP Hardware System Built in Virtualization Leading Performance Processor Innovation Streams Foundations Suzhou PowerCore Technology Virtualization Offerings Key solutions: +Open Source Tools +Middleware +Industry Solutions + Social / Mobile / Analytics / Cloud Hadoop Spark
  10. 10. 10 10 © 2016 International Business Machines Corporation Fundamental forces are accelerating industry change IT innovation can no longer come from just the processor Solution Innovation and Acceleration is a key to the future Price/Performance Full system stack innovation required Moore’s Law Technology and Processors 2000 2020 Firmware / OS Accelerators Software Storage Network Full Stack Acceleratio n (Lower is better) The OpenPOWER Foundation is an open ecosystem, using the POWER Architecture to serve the evolving needs of customers.
  11. 11. 11 11 © 2016 International Business Machines Corporation NVLINK GPUFPGA Flash NIC MRAM PCM Solution Acceleration is a key to the future
  12. 12. 12 12 © 2016 International Business Machines Corporation NVLINK GPU Flash Graphics – CAE - EDA Weather Defense Financial Services Bio-Sciences General: Compression Encryption DataBases: Flash Finance: Algorithms, Facial Genomics : Algorithms Decision Support Data Analytics Financial Simulations Genomic Analysis Network Data Forensics Facial Recognition Solution Acceleration is a key to the future
  13. 13. 13 13 © 2016 International Business Machines Corporation IBM Data Engine for NoSQL is an integrated platform for large and fast growing NoSQL data stores. It builds on the CAPI capability of POWER8 systems and provides super-fast access to large flash storage capacity. It delivers high speed access to both RAM and flash storage which can result in significantly lower cost, and higher workload density for NoSQL deployments than a standard RAM-based system. The solution offers superior performance and price-performance to scale out x86 server deployments that are either limited in available memory per server or have flash memory with limited data access latency. Up to 56TB of extended memory with one POWER8 server + CAPI attach FLASH Power S822L / S812L Flash System 900 Power S822L / S812L / S822 LC NEW External Flash Configuration Integrated Flash Configuration Up to 8TB of super-fast storage tier on one POWER8 server IBM Data Engine for NoSQL Cost Savings for In-Memory NoSQL Data Stores
  14. 14. 14 14 © 2016 International Business Machines Corporation Identical hardware with 3 different paths to data FlashSystem Conventional I/O (FC) CAPI - E IBM POWER S822L CAPI - I IBM's CAPI NVMe Flash Accelerator is almost 5X more efficient in performing IO vs traditional storage. 21% 35% 56% 100% 0% 25% 50% 75% 100% CAPI NVMe Traditional NVMe Traditional Storage - Direct IO Traditional Storage - Filesystem RelativeCAPI vs. NVMe Instruction Counts per IO Kernel Instructions User Instructions ON CAPI Unlocks the Next Level of Performance for Flash
  15. 15. 15 15 © 2016 International Business Machines Corporation ON Efficient IO Enables True Utilization of Storage Bandwidth  Under heavy load, IOPs per thread becomes a critical metric for sustaining throughput in a storage system. As throughput increases, more CPU is required to maintain performance.  CAPI NVMe flash leverages improved path length, architectural improvements, and hardware built-in to POWER8 to greatly- improve the relative IOPs per CPU thread.  At high levels of IO (sustained millions of IOPs), more data can be processed more efficiently, radically changing the amount of CPU required to “feed the (IO) beast.” 0.6X 1X 2.6X 3.7X 0% 100% 200% 300% 400% Fibre Channel NVMe CAPI Fibre Channel CAPI NVMe AverageRelativeIOPs per CPU Thread CAPI-accelerated NVMe Flash can issue 3.7X more IOs per CPU thread than regular NVMe flash.
  16. 16. 16 16 © 2016 International Business Machines Corporation Neo4j + IBM POWER8: Unparalleled Scale and Performance Neo4j on IBM POWER8 The strength and tooling of Neo4j The performance of POWER8 The scalability of POWER8 & CAPI Flash Unrivaled graph application scalability and performance ON
  17. 17. © 2016 IBM Corporation Real-World mixed graph transaction workload running Neo4j on POWER8 delivers 1.82X better performance than Intel Xeon E5-2650 v4 Broadwell 711 390 0 100 200 300 400 500 600 700 800 POWER8 x86 RepresentativemixedworkloadThroughput IBM Power S822LC (20c/160t) x86 Broadwell Server (24c/48t) 82% More Throughput • POWER8 delivers 1.82X more query throughput for a representative mixed sample workload than x86 – POWER8 (20 cores / 256 GB): – x86 system with Broadwell processor (24 cores / 256 GB): •Based on IBM internal testing of single system and OS image running a real-world mixed graph transaction workload based on LDBC benchmark. Conducted under laboratory condition, individual result can vary based on workload size, use of storage subsystems & other conditions. • IBM Power System S822LC; 20 cores (2 x 10c chips) / 160 threads, POWER8; 256 GB memory, Neo4j, Ubuntu 16. Competitive stack: HP Proliant DL380 Gen9; 24 cores (2 x 12c chips) / 48 threads; Intel E5-2650 v4; 256 GB memory, Neo4j, RHEL 7.2 . Pricing is based bundled pricing for S822LC with Integrated CAPI Flash card.
  18. 18. © 2016 International Business Machines Corporation 18 Scale up and/or out based on your application requirements • Out-of-order, super- scalar design for exploiting instruction level parallelization leading to low CPI • Larger caches and 99.94% data-cache hit rate • SMT design to improve core efficiency and increase throughput capability Use the paradigm shift to realize your imagination CAPI-Flash Performance and Scale as YOU Need ON
  19. 19. Open innovation to put data to work across the enterprise Thanks! © 2016 International Business Machines Corporation 19
  20. 20. © Copyright International Business Machines Corporation 2016 Printed in the United States of America September 2016 IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of International Business Machines Corp., registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM or other companies. A current list of IBM trademarks is available on the Web at “Copyright and trademark information” at www.ibm.com/legal/copytrade.shtml. The following terms are trademarks or registered trademarks licensed by Power.org in the United States and/or other countries: Power ISA. Information on the list of U.S. trademarks licensed by Power.org may be found at www.power.org/about/brand-center/. Linux is a trademark of Linus Torvalds in the United States, other countries, or both. Other company, product, and service names may be trademarks or service marks of others. All information contained in this document is subject to change without notice. The products described in this document are NOT intended for use in applications such as implantation, life support, or other hazardous uses where malfunction could result in death, bodily injury, or catastrophic property damage. The information contained in this document does not affect or change IBM product specifications or warranties. Nothing in this document shall operate as an express or implied license or indemnity under the intellectual property rights of IBM or third parties. All information contained in this document was obtained in specific environments, and is presented as an illustration. The results obtained in other operating environments may vary. While the information contained herein is believed to be accurate, such information is preliminary, and should not be relied upon for accuracy or completeness, and no representations or warranties of accuracy or completeness are made. Note: This document contains information on products in the design, sampling and/or initial production phases of development. This information is subject to change without notice. Verify with your IBM field applications engineer that you have the latest version of this document before finalizing a design. You may use this documentation solely for developing technology products compatible with Power Architecture®. You may not modify or distribute this documentation. No license, express or implied, by estoppel or otherwise to any intellectual property rights is granted by this document. THE INFORMATION CONTAINED IN THIS DOCUMENT IS PROVIDED ON AN “AS IS” BASIS. In no event will IBM be liable for damages arising directly or indirectly from any use of the information contained in this document. IBM Systems and Technology Group 2070 Route 52, Bldg. 330 Hopewell Junction, NY 12533-6351 The IBM home page can be found at ibm.com®. Version 1.1 January, 2016

×