Summarizer: Trading Communication with Computing Near Storage
|
|
- Brook Barton
- 5 years ago
- Views:
Transcription
1 Summarizer: Trading Communication with Computing Near Storage Gunjae Koo*, Kiran Kumar Matam*, Te I, H.V. Krishina Giri Nara*, Jing Li, Hung-Wei Tseng, Steven Swanson, Murali Annavaram* *University of Southern California North Carolina State University University of California, San Diego
2 Motivation High Data Movement Cost Host Limited data bandwidth High access latency CPU Storage interface External External (host -- storage) (host -- storage) Internal Internal Data host Data transfer from storage 2
3 Near Data Processing (NDP) Host CPU Storage interface Storage Processor (SP) External (host storage) Internal Data host Data transfer from storage 3
4 Near Data Processing (NDP) Host CPU Storage interface Storage Processor (SP) W/O NDP With NDP Data host External (host storage) Data transfer from storage Data storage Internal 4
5 Near Data Processing (NDP) on SSDs Host CPU Storage interface SP Garbage collection Wearleveling W/O NDP With NDP Data host External (host storage) Data transfer from storage Data computation Data storage Internal 5
6 Near Data Processing (NDP) on SSDs Host Obstacles to in-ssd processing CPU Storage interface SP Less powerful embedded processor Garbage collection Dynamic computation resource availability Wearleveling W/O NDP With NDP External (host storage) Manual workload partitioning is difficult Summarizer: Dynamic NDP framework for SSD Data host Data transfer from storage Data storage Internal 6
7 Summarizer Basic Concept Host Monitoring resources CPU Storage interface AP 7
8 Summarizer Basic Concept Host Monitoring resources CPU Storage interface AP 8
9 Summarizer Detailed Firmware Architecture Storage Interface (PCIe / NVMe) Request queue Response queue Host CPU User Applications / Operating Systems SSD Firmware I/O Controller (NVMe command decoder) Translation Layer (FTL) Summarizer Task Controller SSD Embedded Processors NVMe Host Driver TQ Host Memory SSD SoC Interconnection User Functions SQ CQ Controller DRAM Controller NAND SSD DRAM 9
10 Normal Page Read Request Storage Interface (PCIe / NVMe) Request queue Response queue Host CPU User Applications / Operating Systems SSD Firmware I/O Controller (NVMe command decoder) Translation Layer (FTL) (RD) PPA Summarizer Task Controller NVMe Host Driver RD ( LBA) TQ Host Memory SSD SoC Interconnection User Functions SQ CQ Controller DRAM Controller NAND SSD DRAM 10
11 Normal Page Read Request Storage Interface (PCIe / NVMe) Request queue Response queue Host CPU SSD Firmware Summarizer User Applications / Operating Systems I/O Controller (NVMe command decoder) Translation Layer (FTL) Task Controller NVMe Host Driver RD(PPA 2) RD(PPA 1) TQ Host Memory SSD SoC Interconnection User Functions SQ CQ Controller DRAM Controller NAND Page data SSD DRAM 11
12 Normal Page Read Request Storage Interface (PCIe / NVMe) Request queue Response queue Host CPU User Applications / Operating Systems SSD Firmware I/O Controller (NVMe command decoder) Translation Layer (FTL) Summarizer Task Controller NVMe Host Driver TQ Host Memory SSD SoC Interconnection User Functions SQ CQ Controller DRAM Controller NAND SSD DRAM Page data 12
13 Summarizer Initialization (Function Offloading) Storage Interface (PCIe / NVMe) Request queue Response queue Host CPU User Applications / Operating Systems SSD Firmware I/O Controller (NVMe command decoder) Translation Layer (FTL) Summarizer Task Controller New NVMe command NVMe Host Driver INIT ( foo) TQ Function registration Host foo() Memory SQ CQ SSD SoC Interconnection Controller Function DRAM offloading Controller NAND SSD DRAM User Functions f#1 foo() 13
14 Summarizer Computation (Dynamic mode) Storage Interface (PCIe / NVMe) Request queue Response queue Host CPU User Applications / Operating Systems New NVMe SSD command Firmware decode I/O Controller (NVMe command decoder) Translation Layer (FTL) RD&PROC(PPA,foo) Summarizer Task Controller New NVMe command NVMe Host Driver RD&PROC( LBA,foo) TQ Host Memory SQ CQ SSD SoC Interconnection Controller DRAM Controller NAND SSD DRAM User Functions f#1 foo() f#2 goo() 14
15 Summarizer Computation (Dynamic mode) Storage Interface (PCIe / NVMe) Request queue Response queue Host CPU User Applications / Operating Systems SSD Firmware I/O Controller (NVMe command decoder) Translation Layer (FTL) RD&PROC(PPA,foo) Summarizer Task Controller NVMe Host Driver RD&P(PPA2,foo) RD&P(PPA1,foo) RD&P(PPA1,foo) TQ Host Memory SQ CQ SSD SoC Interconnection Controller DRAM Controller NAND Page data SSD DRAM User Functions f#1 foo() f#2 goo() 15
16 Summarizer Computation (Dynamic mode) Storage Interface (PCIe / NVMe) Request queue Response queue Host CPU User Applications / Operating Systems NVMe Host Driver SSD Firmware I/O Controller (NVMe command decoder) Translation Layer (FTL) RD&PROC(PPA,foo) RD&P(PPA1,foo) CC/Proc Summarizer Task Controller Register in TQ buf1, foo TQ Host Memory SQ CQ SSD SoC Interconnection Controller NAND DRAM Controller SSD DRAM Page data User Functions f#1 foo1() f#2 goo() 16
17 Summarizer Computation (Dynamic mode) Storage Interface (PCIe / NVMe) Request queue Response queue Host CPU User Applications / Operating Systems NVMe Host Driver SSD Firmware I/O Controller (NVMe command decoder) Translation Layer (FTL) RD&PROC(PPA,foo) RD&P(PPA1,foo) CC Summarizer Task Controller TQ is full TQ Host Memory SQ CQ SSD SoC Interconnection Controller NAND DRAM Controller SSD DRAM Page data User Functions f#1 foo() f#2 goo() 17
18 Summarizer Finalization Storage Interface (PCIe / NVMe) Request queue Response queue Host CPU User Applications / Operating Systems SSD Firmware I/O Controller (NVMe command decoder) Translation Layer (FTL) Summarizer Task Controller New NVMe command NVMe Host Driver FINAL ( foo) TQ Host Memory SQ CQ SSD SoC Interconnection Controller NAND DRAM Controller SSD DRAM Results User Functions f#1 foo() f#2 goo() 18
19 Summarizer API and NVMe commands Initialization NVMe command: INIT_TSKn Transfer a in-ssd procedure to SSD memory Initialize data structure and temporal variables for in-ssd computation Computation NVMe command: READ_PROC_TSKn Page read command is issued with the flag indicating the user procedure embedded in SSD memory Return the special code if the requested page is processed in SSD Page data is transferred to the host if the requested page is NOT computed in SSD Finalization NVMe command: FINAL_TSKn Gather final in-ssd computation results and transfer to the host 19
20 PCIe (host LS2085a) PCIe (LS2085a - FPGA) Evaluation Platform LS2085a intelligent SSD development platform ARM cores running FTL and Summarizerfirmware FPGA implementing NAND flash controller PCIe Gen. 3 4x lanes for host communication LS2085a CPU CPU CPU CPU L1D (32KB) L1I (48KB) L1D (32KB) L1I (48KB) L1D (32KB) L1I (48KB) L1D (32KB) L1I (48KB) L2 (1MB) Interconnection L2 (1MB) FPGA (ALTERA Stratix V) DDR4 Memory Controller DRAM DRAM NAND NAND flash flash DIMM DIMMs 20
21 Evaluation - Performance 4 SDD time TPC-H Query6 Host time Static workload offloading Static Dynamic 21
22 Evaluation - Performance 4 SDD time TPC-H Query6 Host time CPU only processing (baseline) SSD only processing Static Dynamic 22
23 Evaluation - Performance 4 SDD time TPC-H Query6 Host time Summarizer Dynamic Offloading Static Dynamic 23
24 Evaluation - Performance 4 SDD time TPC-H Query6 Host time SSD processing + transfer time 3 (internal + external + In-SSD processing) Host CPU processing time Static Dynamic 24
25 Evaluation - Performance 4 TPC-H Query6 Execution time normalized to baseline (CPU only) SDD time Host time Static Dynamic 25
26 Execution time (normalized to baseline) Evaluation - Performance 4 SDD time TPC-H Query6 Host time Static Dynamic 26
27 Evaluation - Performance Execution time (normalized to baseline) SDD time 0.30 Chart Title SDD TPC-H timequery6 Host time Host time CPU only Static Dynamic Dynamic 27
28 Evaluation - Performance W/O NDP With NDP SDD time Host time Chart Title SDD TPC-H timequery6 Host time Performance improved by 14% 0.24 External (host storage) Data host Data transfer from storage 0.62 Data storage Internal CPU only Static Dynamic Dynamic 28
29 Evaluation - Performance 4 SDD time TPC-H Query6 Host time 3 Performance degraded by static NDP Static Dynamic 29
30 Evaluation - Performance Execution time (normalized to baseline) Execution time (normalized to baseline) Execution time (normalized to baseline) Execution time (normalized to baseline) 16% 10% 20% 7% 30
31 Design Exploration Higher Internal Bandwidth Host Data transfer bottleneck CPU Storage interface Commercial SSD maintains internal bandwidth external bandwidth 31
32 Design Exploration Higher Internal Bandwidth Host Data transfer bottleneck CPU Storage interface SP Higher internal bandwidth without increasing external bandwidth 32
33 Design Exploration Higher Internal Bandwidth Speedup 100% 80% 60% 40% External : Internal bandwidth ratio 20% 0% 1:1 1:2 1:3 1:4 1:1 1:2 1:3 1:4 1:1 1:2 1:3 1:4 1:1 1:2 1:3 1:4 1:1 1:2 1:3 1:4 TPC-H Query 6 TPC-H Query 1 TPC-H Query 14 String Similarity Join Average 33
34 Design Exploration Higher Internal Bandwidth Speedup Summarizer is effective if an SSD platform has higher internal bandwidth 100% 80% 60% 40% 20% 0% 1:1 1:2 1:3 1:4 1:1 1:2 1:3 1:4 1:1 1:2 1:3 1:4 1:1 1:2 1:3 1:4 1:1 1:2 1:3 1:4 TPC-H Query 6 TPC-H Query 1 TPC-H Query 14 String Similarity Join Average 34
35 Design Exploration Better SSD Processor Host CPU Storage interface AP Better embedded processor is cost effective 35
36 Design Exploration Higher Internal Bandwidth Speedup 120% 100% 80% 60% 40% Embedded processor performance 20% 0% X1 X2 X4 X8 X16 X1 X2 X4 X8 X16 X1 X2 X4 X8 X16 X1 X2 X4 X8 X16 X1 X2 X4 X8 X16 TPC-H Query6 TPC-H Query1 TPC-H Query14 String Similarity Join Average 36
37 Design Exploration Higher Internal Bandwidth Speedup Summarizer is a cost effective NDP solution with powerful storage processors 120% 100% 80% 60% 40% 20% 0% X1 X2 X4 X8 X16 X1 X2 X4 X8 X16 X1 X2 X4 X8 X16 X1 X2 X4 X8 X16 X1 X2 X4 X8 X16 TPC-H Query6 TPC-H Query1 TPC-H Query14 String Similarity Join Average 37
38 Conclusion Dynamic computation offloading framework Dynamic NDP framework for SSDs Opportunistic in-ssd computation Page-level task control Optimal performance improvement Opportunistically enables in-ssd processing Page-level NDP control Automatic workload partitioning Summrizer programming model Summarizer programming model Evaluation on the real development platform Explored design space for future SSDs 38
39 Thank you Summarizer: Trading Communication with Computing Near Storage Gunjae Koo, Kiran Kumar Matam, Te I, H. V. Krishna Giri Nara, Jing Li, Hung-Wei Tseng, Steven Swanson, Murali Annavaram (We thank to Dell EMC for supporting the SSD development board)
Hung-Wei Tseng. Assistant Professor
Hung-Wei Tseng Assistant Professor 919-515-7354 Dept. of Computer Science hungwei tseng@ncsu.edu North Carolina State University http://people.engr.ncsu.edu/htseng3 Raleigh, NC 27695-8206 Education University
More informationI N V E N T I V E. SSD Firmware Complexities and Benefits from NVMe. Steven Shrader
I N V E N T I V E SSD Firmware Complexities and Benefits from NVMe Steven Shrader Agenda Introduction NVMe architectural issues from NVMe functions Structures to model the problem Methods (metadata attributes)
More informationMoneta: A High-Performance Storage Architecture for Next-generation, Non-volatile Memories
Moneta: A High-Performance Storage Architecture for Next-generation, Non-volatile Memories Adrian M. Caulfield Arup De, Joel Coburn, Todor I. Mollov, Rajesh K. Gupta, Steven Swanson Non-Volatile Systems
More informationSoftFlash: Programmable Storage in Future Data Centers Jae Do Researcher, Microsoft Research
SoftFlash: Programmable Storage in Future Data Centers Jae Do Researcher, Microsoft Research 1 The world s most valuable resource Data is everywhere! May. 2017 Values from Data! Need infrastructures for
More informationNear-Data Processing for Differentiable Machine Learning Models
Near-Data Processing for Differentiable Machine Learning Models Hyeokjun Choe 1, Seil Lee 1, Hyunha Nam 1, Seongsik Park 1, Seijoon Kim 1, Eui-Young Chung 2 and Sungroh Yoon 1,3 1 Electrical and Computer
More informationOnyx: A Prototype Phase-Change Memory Storage Array
Onyx: A Prototype Phase-Change Memory Storage Array Ameen Akel * Adrian Caulfield, Todor Mollov, Rajesh Gupta, Steven Swanson Non-Volatile Systems Laboratory, Department of Computer Science and Engineering
More informationExploring System Challenges of Ultra-Low Latency Solid State Drives
Exploring System Challenges of Ultra-Low Latency Solid State Drives Sungjoon Koh Changrim Lee, Miryeong Kwon, and Myoungsoo Jung Computer Architecture and Memory systems Lab Executive Summary Motivation.
More informationDeveloping Low Latency NVMe Systems for HyperscaleData Centers. Prepared by Engling Yeo Santa Clara, CA Date: 08/04/2017
Developing Low Latency NVMe Systems for HyperscaleData Centers Prepared by Engling Yeo Santa Clara, CA 95054 Date: 08/04/2017 Quality of Service IOPS, Throughput, Latency Short predictable read latencies
More informationN V M e o v e r F a b r i c s -
N V M e o v e r F a b r i c s - H i g h p e r f o r m a n c e S S D s n e t w o r k e d f o r c o m p o s a b l e i n f r a s t r u c t u r e Rob Davis, VP Storage Technology, Mellanox OCP Evolution Server
More informationNear- Data Computa.on: It s Not (Just) About Performance
Near- Data Computa.on: It s Not (Just) About Performance Steven Swanson Non- Vola0le Systems Laboratory Computer Science and Engineering University of California, San Diego 1 Solid State Memories NAND
More informationMQSim: A Framework for Enabling Realistic Studies of Modern Multi-Queue SSD Devices
MQSim: A Framework for Enabling Realistic Studies of Modern Multi-Queue SSD Devices Arash Tavakkol, Juan Gómez-Luna, Mohammad Sadrosadati, Saugata Ghose, Onur Mutlu February 13, 2018 Executive Summary
More information2017 Storage Developer Conference. Mellanox Technologies. All Rights Reserved.
Ethernet Storage Fabrics Using RDMA with Fast NVMe-oF Storage to Reduce Latency and Improve Efficiency Kevin Deierling & Idan Burstein Mellanox Technologies 1 Storage Media Technology Storage Media Access
More informationReducing Solid-State Storage Device Write Stress Through Opportunistic In-Place Delta Compression
Reducing Solid-State Storage Device Write Stress Through Opportunistic In-Place Delta Compression Xuebin Zhang, Jiangpeng Li, Hao Wang, Kai Zhao and Tong Zhang xuebinzhang.rpi@gmail.com ECSE Department,
More informationAccelerating Microsoft SQL Server Performance With NVDIMM-N on Dell EMC PowerEdge R740
Accelerating Microsoft SQL Server Performance With NVDIMM-N on Dell EMC PowerEdge R740 A performance study with NVDIMM-N Dell EMC Engineering September 2017 A Dell EMC document category Revisions Date
More informationAccelerating Data Centers Using NVMe and CUDA
Accelerating Data Centers Using NVMe and CUDA Stephen Bates, PhD Technical Director, CSTO, PMC-Sierra Santa Clara, CA 1 Project Donard @ PMC-Sierra Donard is a PMC CTO project that leverages NVM Express
More informationReplacing the FTL with Cooperative Flash Management
Replacing the FTL with Cooperative Flash Management Mike Jadon Radian Memory Systems www.radianmemory.com Flash Memory Summit 2015 Santa Clara, CA 1 Data Center Primary Storage WORM General Purpose RDBMS
More informationUnblinding the OS to Optimize User-Perceived Flash SSD Latency
Unblinding the OS to Optimize User-Perceived Flash SSD Latency Woong Shin *, Jaehyun Park **, Heon Y. Yeom * * Seoul National University ** Arizona State University USENIX HotStorage 2016 Jun. 21, 2016
More informationManaging Array of SSDs When the Storage Device is No Longer the Performance Bottleneck
Managing Array of Ds When the torage Device is No Longer the Performance Bottleneck Byung. Kim, Jaeho Kim, am H. Noh UNIT (Ulsan National Institute of cience & Technology) Outline Motivation & Observation
More informationMorpheus: Creating Application Objects Efficiently for Heterogeneous Computing
Morpheus: Creating Application Objects Efficiently for Heterogeneous Computing Hung-Wei Tseng, Qianchen Zhao, Yuxiao Zhou, Mark Gahagan, Steven Swanson Department of Computer Science and Engineering University
More informationLinux Storage System Bottleneck Exploration
Linux Storage System Bottleneck Exploration Bean Huo / Zoltan Szubbocsev Beanhuo@micron.com / zszubbocsev@micron.com 215 Micron Technology, Inc. All rights reserved. Information, products, and/or specifications
More informationMorpheus: Creating Application Objects Efficiently for Heterogeneous Computing
216 ACM/IEEE 43rd Aual International Symposium on Computer Architecture Morpheus: Creating Application Objects Efficiently for Heterogeneous Computing Hung-Wei Tseng, Qianchen Zhao, Yuxiao Zhou, Mark Gahagan,
More informationOn the Portability and Performance of Message-Passing Programs on Embedded Multicore Platforms
On the Portability and Performance of Message-Passing Programs on Embedded Multicore Platforms Shih-Hao Hung, Po-Hsun Chiu, Chia-Heng Tu, Wei-Ting Chou and Wen-Long Yang Graduate Institute of Networking
More informationBeyond Block I/O: Rethinking
Beyond Block I/O: Rethinking Traditional Storage Primitives Xiangyong Ouyang *, David Nellans, Robert Wipfel, David idflynn, D. K. Panda * * The Ohio State University Fusion io Agenda Introduction and
More informationWarped-Preexecution: A GPU Pre-execution Approach for Improving Latency Hiding
Warped-Preexecution: A GPU Pre-execution Approach for Improving Latency Hiding Keunsoo Kim, Sangpil Lee, Myung Kuk Yoon, *Gunjae Koo, Won Woo Ro, *Murali Annavaram Yonsei University *University of Southern
More informationPresented by: Nafiseh Mahmoudi Spring 2017
Presented by: Nafiseh Mahmoudi Spring 2017 Authors: Publication: Type: ACM Transactions on Storage (TOS), 2016 Research Paper 2 High speed data processing demands high storage I/O performance. Flash memory
More informationA Caching-Oriented FTL Design for Multi-Chipped Solid-State Disks. Yuan-Hao Chang, Wei-Lun Lu, Po-Chun Huang, Lue-Jane Lee, and Tei-Wei Kuo
A Caching-Oriented FTL Design for Multi-Chipped Solid-State Disks Yuan-Hao Chang, Wei-Lun Lu, Po-Chun Huang, Lue-Jane Lee, and Tei-Wei Kuo 1 June 4, 2011 2 Outline Introduction System Architecture A Multi-Chipped
More informationWillow: A User- Programmable SSD
Willow: A User- Programmable SSD Sudharsan Seshadri, Mark Gahagan, Sundaram Bhaskaran, Trevor Bunker, Arup De, Yanqin Jin, Yang Liu, and Steven Swanson Non- VolaDle Systems Laboratory Computer Science
More informationTransparent Offloading and Mapping (TOM) Enabling Programmer-Transparent Near-Data Processing in GPU Systems Kevin Hsieh
Transparent Offloading and Mapping () Enabling Programmer-Transparent Near-Data Processing in GPU Systems Kevin Hsieh Eiman Ebrahimi, Gwangsun Kim, Niladrish Chatterjee, Mike O Connor, Nandita Vijaykumar,
More informationAn FPGA Based Enterprise SSD Reference Design. Amit Saxena, VP, Engineering. The IP enabled solutions provider
An FPGA Based Enterprise SSD Reference Design Amit Saxena, VP, Engineering The IP enabled solutions provider AGENDA FPGA Based Enterprise SSDC 8/14/2015 2015 Copyright Mobiveil Inc. 2 FPGA Based NVMe Enterprise
More informationEmerging NVM Features
Emerging NVM Features For Emerging NVM Interface (I/F)s Thomas Won Ha Choi SK hynix Santa Clara, CA 1 Introductions Thomas Won Ha Choi Senior Engineer, SK hynix DRAM Server Product Planning Specialties:
More informationPartitioned Real-Time NAND Flash Storage. Katherine Missimer and Rich West
Partitioned Real-Time NAND Flash Storage Katherine Missimer and Rich West Introduction Eric Risberg AP CircuitsToday 2 Introduction Eric Risberg AP CircuitsToday Analytics Vidhya 3 Chesky_W Mapping Ignorance
More informationDesign Choices for FPGA-based SoCs When Adding a SATA Storage }
U4 U7 U7 Q D U5 Q D Design Choices for FPGA-based SoCs When Adding a SATA Storage } Lorenz Kolb & Endric Schubert, Missing Link Electronics Rudolf Usselmann, ASICS World Services Motivation for SATA Storage
More informationGen-Z Overview. 1. Introduction. 2. Background. 3. A better way to access data. 4. Why a memory-semantic fabric
Gen-Z Overview 1. Introduction Gen-Z is a new data access technology that will allow business and technology leaders, to overcome current challenges with the existing computer architecture and provide
More informationTen Ways to Improve Flash Storage System Performance
Ten Ways to Improve Flash Storage System Performance Camberley Bates, Evaluator Group @camberleyb Panel Moderator August 2018 1 Abstract Quite frequently, the wonderful flash storage systems, despite having
More informationXPU A Programmable FPGA Accelerator for Diverse Workloads
XPU A Programmable FPGA Accelerator for Diverse Workloads Jian Ouyang, 1 (ouyangjian@baidu.com) Ephrem Wu, 2 Jing Wang, 1 Yupeng Li, 1 Hanlin Xie 1 1 Baidu, Inc. 2 Xilinx Outlines Background - FPGA for
More informationEnabling NVMe I/O Scale
Enabling NVMe I/O Determinism @ Scale Chris Petersen, Hardware System Technologist Wei Zhang, Software Engineer Alexei Naberezhnov, Software Engineer Facebook Facebook @ Scale 800 Million 1.3 Billion 2.2
More informationMoneta: A High-performance Storage Array Architecture for Nextgeneration, Micro 2010
Moneta: A High-performance Storage Array Architecture for Nextgeneration, Non-volatile Memories Micro 2010 NVM-based SSD NVMs are replacing spinning-disks Performance of disks has lagged NAND flash showed
More informationAltera SDK for OpenCL
Altera SDK for OpenCL A novel SDK that opens up the world of FPGAs to today s developers Altera Technology Roadshow 2013 Today s News Altera today announces its SDK for OpenCL Altera Joins Khronos Group
More informationAn NVMe-based FPGA Storage Workload Accelerator
An NVMe-based FPGA Storage Workload Accelerator Dr. Sean Gibb, VP Software Eideticom Santa Clara, CA 1 PCIe Bus NVMe SSD NVMe SSD Acceleration Host CPU HDD RDMA NIC NoLoad Accel. Card TM Storage I/O Bandwidth
More informationFlashShare: Punching Through Server Storage Stack from Kernel to Firmware for Ultra-Low Latency SSDs
FlashShare: Punching Through Server Storage Stack from Kernel to Firmware for Ultra-Low Latency SSDs Jie Zhang, Miryeong Kwon, Donghyun Gouk, Sungjoon Koh, Changlim Lee, Mohammad Alian, Myoungjun Chun,
More informationBringing Intelligence to Enterprise Storage Drives
Bringing Intelligence to Enterprise Storage Drives Neil Werdmuller Director Storage Solutions Arm Santa Clara, CA 1 Who am I? 28 years experience in embedded Lead the storage solutions team Work closely
More informationComputational Storage: Acceleration Through Intelligence & Agility
Flash Memory Summit Computational Storage: Acceleration Through Intelligence & Agility Dr. Hao Zhong CEO & Co-Founder, ScaleFlux Flash Memory Summit 2018 Santa Clara, CA What s the Big Deal? High Cost
More informationLightweight KV-based Distributed Store for Datacenters
Lightweight KV-based Distributed Store for Datacenters Chanwoo Chung, Jinhyung Koo*, Arvind, and Sungjin Lee Massachusetts Institute of Technology (MIT) Daegu Gyeongbuk Institute of Science & Technology
More informationPCIe Storage Beyond SSDs
PCIe Storage Beyond SSDs Fabian Trumper NVM Solutions Group PMC-Sierra Santa Clara, CA 1 Classic Memory / Storage Hierarchy FAST, VOLATILE CPU Cache DRAM Performance Gap Performance Tier (SSDs) SLOW, NON-VOLATILE
More informationEnabling the NVMe CMB and PMR Ecosystem
Architected for Performance Enabling the NVMe CMB and PMR Ecosystem Stephen Bates, PhD. CTO, Eideticom Oren Duer. Software Architect, Mellanox NVM Express Developers Day May 1, 2018 Outline 1. Intro to
More informationBringing Intelligence to Enterprise Storage Drives
Bringing Intelligence to Enterprise Storage Drives Neil Werdmuller Director Storage Solutions Arm Santa Clara, CA 1 Who am I? 28 years experience in embedded Lead the storage solutions team Work closely
More informationTRADITIONAL search engines utilize hard disk drives
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 1.119/TC.216.268818,
More informationUsing Transparent Compression to Improve SSD-based I/O Caches
Using Transparent Compression to Improve SSD-based I/O Caches Thanos Makatos, Yannis Klonatos, Manolis Marazakis, Michail D. Flouris, and Angelos Bilas {mcatos,klonatos,maraz,flouris,bilas}@ics.forth.gr
More informationProgrammable Solutions for Data Center Applications
Programmable Solutions for Data Center Applications DS McIntyre Consulting dmm961@gmail.com 1 Topics Data Center Trends o Storage, Compute, Networking Technology Options FPGA Examples 2 Data Center Macro
More informationCatapult: A Reconfigurable Fabric for Petaflop Computing in the Cloud
Catapult: A Reconfigurable Fabric for Petaflop Computing in the Cloud Doug Burger Director, Hardware, Devices, & Experiences MSR NExT November 15, 2015 The Cloud is a Growing Disruptor for HPC Moore s
More informationHardware NVMe implementation on cache and storage systems
Hardware NVMe implementation on cache and storage systems Jerome Gaysse, IP-Maker Santa Clara, CA 1 Agenda Hardware architecture NVMe for storage NVMe for cache/application accelerator NVMe for new NVM
More informationTechnology Advancement in SSDs and Related Ecosystem Changes
Technology Advancement in SSDs and Related Ecosystem Changes Sanjeev Kumar/ Ravish Sharma Software Product Engineering, HiTech, Tata Consultancy Services 27 May 2016 1 SDC India 2016 Agenda Disruptive
More informationOpen-Channel SSDs Offer the Flexibility Required by Hyperscale Infrastructure Matias Bjørling CNEX Labs
Open-Channel SSDs Offer the Flexibility Required by Hyperscale Infrastructure Matias Bjørling CNEX Labs 1 Public and Private Cloud Providers 2 Workloads and Applications Multi-Tenancy Databases Instance
More informationToward a Memory-centric Architecture
Toward a Memory-centric Architecture Martin Fink EVP & Chief Technology Officer Western Digital Corporation August 8, 2017 1 SAFE HARBOR DISCLAIMERS Forward-Looking Statements This presentation contains
More informationThe Impact of SSD Selection on SQL Server Performance. Solution Brief. Understanding the differences in NVMe and SATA SSD throughput
Solution Brief The Impact of SSD Selection on SQL Server Performance Understanding the differences in NVMe and SATA SSD throughput 2018, Cloud Evolutions Data gathered by Cloud Evolutions. All product
More informationAccelerating Real-Time Big Data. Breaking the limitations of captive NVMe storage
Accelerating Real-Time Big Data Breaking the limitations of captive NVMe storage 18M IOPs in 2u Agenda Everything related to storage is changing! The 3rd Platform NVM Express architected for solid state
More informationAn Adaptive Partitioning Scheme for DRAM-based Cache in Solid State Drives
An Adaptive Partitioning Scheme for DRAM-based Cache in Solid State Drives Hyotaek Shimy, Bon-Keun Seoy, Jin-Soo Kimz, and Seungryoul Maeng Computer Science Department, Korea Advanced Institute of Science
More informationFC-NVMe. NVMe over Fabrics. Fibre Channel the most trusted fabric can transport NVMe natively. White Paper
FC-NVMe NVMe over Fabrics Fibre Channel the most trusted fabric can transport NVMe natively BACKGROUND AND SUMMARY Ever since IBM shipped the world s first hard disk drive (HDD), the RAMAC 305 in 1956,
More informationNVMe Direct. Next-Generation Offload Technology. White Paper
NVMe Direct Next-Generation Offload Technology The market introduction of high-speed NVMe SSDs and 25/40/50/100Gb Ethernet creates exciting new opportunities for external storage NVMe Direct enables high-performance
More informationOnyx: A Protoype Phase Change Memory Storage Array
Onyx: A Protoype Phase Change ory Storage Array Ameen Akel Adrian M. Caulfield Todor I. Mollov Rajesh K. Gupta Steven Swanson Computer Science and Engineering University of California, San Diego Abstract
More informationEvolution of Rack Scale Architecture Storage
Evolution of Rack Scale Architecture Storage Murugasamy (Sammy) Nachimuthu, Principal Engineer Mohan J Kumar, Fellow Intel Corporation August 2016 1 Agenda Introduction to Intel Rack Scale Design Storage
More informationGreen Memory Solution. Jung-Bae Lee
Green Memory Solution Jung-Bae Lee Contents Introduction Data Explosion and Data Center Expansion Data Center Issues: Power & Performance Samsung Green Memory Solution History of Green Memory Memory &
More informationAccelerate Applications Using EqualLogic Arrays with directcache
Accelerate Applications Using EqualLogic Arrays with directcache Abstract This paper demonstrates how combining Fusion iomemory products with directcache software in host servers significantly improves
More informationHP SSD EX920 M.2. 2TB Sustained sequential read: Up to 3200 MB/s Sustained sequential write: Up to 1600 MB/s
HP SSD EX920 M.2 Product Specification Capacity: 256GB, 512GB, 1TB, 2TB Components: 3D NAND/ DRAM Cache Read and Write IOPS (Iometer* Queue Depth 32) 256 GB Random 4 KB reads: Up to 180K IOPS Random 4
More informationSamsung PM1725a NVMe SSD
Samsung PM1725a NVMe SSD Exceptionally fast speeds and ultra-low latency for enterprise application Brochure 1 Extreme performance from an SSD technology leader Maximize data transfer with the high-performance,
More informationmemory VT-PM8 & VT-PM16 EVALUATION WHITEPAPER Persistent Memory Dual Port Persistent Memory with Unlimited DWPD Endurance
memory WHITEPAPER Persistent Memory VT-PM8 & VT-PM16 EVALUATION VT-PM drives, part of Viking s persistent memory technology family of products, are 2.5 U.2 NVMe PCIe Gen3 drives optimized with Radian Memory
More informationNVMe SSD s. NVMe is displacing SATA in applications which require performance. NVMe has excellent programing model for host software
NVMe SSD s NVMe is displacing SATA in applications which require performance NVMe has excellent programing model for host software Latency is becoming the key driving force for system performance, although
More informationFMS18 Invited Session 101-B1 Hardware Acceleration Techniques for NVMe-over-Fabric
Flash Memory Summit 2018 Santa Clara, CA FMS18 Invited Session 101-B1 Hardware Acceleration Techniques for NVMe-over-Fabric Paper Abstract: The move from direct-attach to Composable Infrastructure is being
More informationHow Are The Networks Coping Up With Flash Storage
How Are The Networks Coping Up With Flash Storage Saurabh Sureka Sr. Product Manager, Emulex, an Avago Technologies Company www.emulex.com Santa Clara, CA 1 Goals Data deluge quick peek The flash landscape
More informationSDA: Software-Defined Accelerator for general-purpose big data analysis system
SDA: Software-Defined Accelerator for general-purpose big data analysis system Jian Ouyang(ouyangjian@baidu.com), Wei Qi, Yong Wang, Yichen Tu, Jing Wang, Bowen Jia Baidu is beyond a search engine Search
More informationEnd-to-End Adaptive Packet Aggregation for High-Throughput I/O Bus Network Using Ethernet
Hot Interconnects 2014 End-to-End Adaptive Packet Aggregation for High-Throughput I/O Bus Network Using Ethernet Green Platform Research Laboratories, NEC, Japan J. Suzuki, Y. Hayashi, M. Kan, S. Miyakawa,
More informationDCS-ctrl: A Fast and Flexible Device-Control Mechanism for Device-Centric Server Architecture
DCS-ctrl: A Fast and Flexible ice-control Mechanism for ice-centric Server Architecture Dongup Kwon 1, Jaehyung Ahn 2, Dongju Chae 2, Mohammadamin Ajdari 2, Jaewon Lee 1, Suheon Bae 1, Youngsok Kim 1,
More informationSolid State Storage is Everywhere Where Does it Work Best?
Solid State Storage is Everywhere Where Does it Work Best? Dennis Martin, President, Demartek www.storagedecisions.com Agenda Demartek About Us Solid-state storage overview Different places to deploy SSD
More informationHewlett Packard Enterprise HPE GEN10 PERSISTENT MEMORY PERFORMANCE THROUGH PERSISTENCE
Hewlett Packard Enterprise HPE GEN10 PERSISTENT MEMORY PERFORMANCE THROUGH PERSISTENCE Digital transformation is taking place in businesses of all sizes Big Data and Analytics Mobility Internet of Things
More information3ME2 Series. Customer Approver. Innodisk Approver. Customer: Customer Part Number: Innodisk Part Number: Innodisk Model Name: Date:
3ME2 Series Customer: Customer Part Number: Innodisk Part Number: Innodisk Model Name: Date: Innodisk Approver Customer Approver Table of contents LIST OF FIGURES... 6 1. PRODUCT OVERVIEW... 7 1.1 INTRODUCTION
More informationThe Transition to PCI Express* for Client SSDs
The Transition to PCI Express* for Client SSDs Amber Huffman Senior Principal Engineer Intel Santa Clara, CA 1 *Other names and brands may be claimed as the property of others. Legal Notices and Disclaimers
More informationSamsung Z-SSD SZ985. Ultra-low Latency SSD for Enterprise and Data Centers. Brochure
Samsung Z-SSD SZ985 Ultra-low Latency SSD for Enterprise and Data Centers Brochure 1 A high-speed storage device from the SSD technology leader Samsung Z-SSD SZ985 offers more capacity than PRAM-based
More informationOpen Channel Solid State Drives NVMe Specification
Open Channel Solid State Drives NVMe Specification Revision 1.2 April 2016 Please write to Matias at mb@lightnvm.io for collaboration Table of Contents 1. Introduction 1.1 Definitions 1.1.1 physical media
More informationImproving DRAM Performance by Parallelizing Refreshes with Accesses
Improving DRAM Performance by Parallelizing Refreshes with Accesses Kevin Chang Donghyuk Lee, Zeshan Chishti, Alaa Alameldeen, Chris Wilkerson, Yoongu Kim, Onur Mutlu Executive Summary DRAM refresh interferes
More informationDeploy a High-Performance Database Solution: Cisco UCS B420 M4 Blade Server with Fusion iomemory PX600 Using Oracle Database 12c
White Paper Deploy a High-Performance Database Solution: Cisco UCS B420 M4 Blade Server with Fusion iomemory PX600 Using Oracle Database 12c What You Will Learn This document demonstrates the benefits
More informationGen-Z Memory-Driven Computing
Gen-Z Memory-Driven Computing Our vision for the future of computing Patrick Demichel Distinguished Technologist Explosive growth of data More Data Need answers FAST! Value of Analyzed Data 2005 0.1ZB
More informationA Semi Preemptive Garbage Collector for Solid State Drives. Junghee Lee, Youngjae Kim, Galen M. Shipman, Sarp Oral, Feiyi Wang, and Jongman Kim
A Semi Preemptive Garbage Collector for Solid State Drives Junghee Lee, Youngjae Kim, Galen M. Shipman, Sarp Oral, Feiyi Wang, and Jongman Kim Presented by Junghee Lee High Performance Storage Systems
More informationAn NVMe-based Offload Engine for Storage Acceleration Sean Gibb, Eideticom Stephen Bates, Raithlin
An NVMe-based Offload Engine for Storage Acceleration Sean Gibb, Eideticom Stephen Bates, Raithlin 1 Overview Acceleration for Storage NVMe for Acceleration How are we using (abusing ;-)) NVMe to support
More informationMass-Storage Structure
Operating Systems (Fall/Winter 2018) Mass-Storage Structure Yajin Zhou (http://yajin.org) Zhejiang University Acknowledgement: some pages are based on the slides from Zhi Wang(fsu). Review On-disk structure
More informationStorage Systems : Disks and SSDs. Manu Awasthi July 6 th 2018 Computer Architecture Summer School 2018
Storage Systems : Disks and SSDs Manu Awasthi July 6 th 2018 Computer Architecture Summer School 2018 Why study storage? Scalable High Performance Main Memory System Using Phase-Change Memory Technology,
More informationMemory Expansion Technology Using Software-Controlled SSD
Memory Expansion Technology Using Software-Controlled SSD S. Kazama*, S. Gokita*, S. Kuwamura*, E. Yoshida*, J. Ogawa*, Y. Honda** *Fujitsu Laboratories Ltd. **Fujitsu Ltd. Contact: sc-ssd-fms2017@ml.labs.fujitsu.com
More informationAccelerating Storage with NVM Express SSDs and P2PDMA Stephen Bates, PhD Chief Technology Officer
Accelerating Storage with NVM Express SSDs and P2PDMA Stephen Bates, PhD Chief Technology Officer 2018 Storage Developer Conference. Eidetic Communications Inc. All Rights Reserved. 1 Outline Motivation
More informationLow-Overhead Flash Disaggregation via NVMe-over-Fabrics Vijay Balakrishnan Memory Solutions Lab. Samsung Semiconductor, Inc.
Low-Overhead Flash Disaggregation via NVMe-over-Fabrics Vijay Balakrishnan Memory Solutions Lab. Samsung Semiconductor, Inc. 1 DISCLAIMER This presentation and/or accompanying oral statements by Samsung
More informationOSSD: A Case for Object-based Solid State Drives
MSST 2013 2013/5/10 OSSD: A Case for Object-based Solid State Drives Young-Sik Lee Sang-Hoon Kim, Seungryoul Maeng, KAIST Jaesoo Lee, Chanik Park, Samsung Jin-Soo Kim, Sungkyunkwan Univ. SSD Desktop Laptop
More informationgenzconsortium.org Gen-Z Technology: Enabling Memory Centric Architecture
Gen-Z Technology: Enabling Memory Centric Architecture Why Gen-Z? Gen-Z Consortium 2017 2 Why Gen-Z? Gen-Z Consortium 2017 3 Why Gen-Z? Businesses Need to Monetize Data Big Data AI Machine Learning Deep
More informationRedrawing the Boundary Between So3ware and Storage for Fast Non- Vola;le Memories
Redrawing the Boundary Between So3ware and Storage for Fast Non- Vola;le Memories Steven Swanson Director, Non- Vola;le System Laboratory Computer Science and Engineering University of California, San
More informationSHRD: Improving Spatial Locality in Flash Storage Accesses by Sequentializing in Host and Randomizing in Device
SHRD: Improving Spatial Locality in Flash Storage Accesses by Sequentializing in Host and Randomizing in Device Hyukjoong Kim 1, Dongkun Shin 1, Yun Ho Jeong 2 and Kyung Ho Kim 2 1 Samsung Electronics
More informationEnyx soft-hardware design services and development framework for FPGA & SoC
soft-hardware design services and development framework for FPGA & SoC Smart NIC Smart Switch Your custom hardware hardware acceleration experts 3rd party IP Cores AXI ARM DMA CPU Your own soft-hardware
More informationNVM Express 1.3 Delivering Continuous Innovation
Architected for Performance NVM Express 1.3 Delivering Continuous Innovation June 2017 Jonmichael Hands, Product Marketing Manager Intel, NVM Express Marketing Co-Chair View recorded webcast NVMe 1.3 -
More informationFlash Memory. SATA SSD vs. PCIe NVMe SSD. White Paper F-WP003
Flash Memory SATA SSD vs. NVMe SSD White Paper F-WP003 Corporate Headquarters: 39870 Eureka Dr., Newark, CA 94560, USA Tel: (510) 623-1231 Fax: (510) 623-1434 E-mail: info@smartm.com Customer Service:
More informationNext Generation Architecture for NVM Express SSD
Next Generation Architecture for NVM Express SSD Dan Mahoney CEO Fastor Systems Copyright 2014, PCI-SIG, All Rights Reserved 1 NVMExpress Key Characteristics Highest performance, lowest latency SSD interface
More informationMulti-Host Sharing of NVMe Drives and GPUs Using PCIe Fabrics
Multi-Host Sharing of NVMe Drives and s Using PCIe Fabrics Vincent Haché Principal Applications Engineer, Microsemi Corporation Santa Clara, CA 1 Introduction Increase in use of s and NVM in DC System
More informationASIC/Merchant Silicon Chip-Based Flash Controllers
ASIC/erchant Silicon Chip-Based Flash Controllers Jeff Yang Silicon otion Flash emory Summit 27 Santa Clara, CA Basic architecture TC region SC region Write channel Read channel Buffer for DATA CP Buffer
More informationPerformance Assessment of an All-RRAM Solid State Drive Through a Cloud-Based Simulation Framework
Performance Assessment of an All-RRAM Solid State Drive Through a Cloud-Based Simulation Framework Lorenzo Zuolo* Michele Cirella, Cristian Zambelli, Rino Micheloni*, and Piero Olivo Lorenzo Zuolo, Michele
More informationMemory Systems DRAM, etc.
Memory Systems DRAM, etc. Prof. Bruce Jacob Keystone Professor & Director of Computer Engineering Program Electrical & Computer Engineering University of Maryland at College Park Today s Story DRAM (the
More information