YouChoose: A Performance Interface Enabling Convenient and Efficient QoS Support for Consolidated Storage Systems
|
|
- Toby Wilkinson
- 5 years ago
- Views:
Transcription
1 YouChoose: A Performance Interface Enabling Convenient and Efficient QoS Support for Consolidated Storage Systems Xuechen Zhang Yuhai Xu Song Jiang The ECE Department Wayne State University Detroit, MI The 27th IEEE Symposium on Massive Storage Systems and Technologies(MSST '11)
2 Outline Introduction Related Work Proposed Scheme Performance Evaluation Conclusions
3 Introduction (1/5) Applications become more data intensive Scientific applications may analyze large data sets Internet search and E-commerce rely on efficient data access Applications performance highly depends on I/O service quality Advantages of consolidated storage system High utilization due to resource sharing Cost-effectiveness of centralized management Low operating cost however, lack of an effective performance interface through which users can present their goals on quality of I/O service (QoS)
4 Introduction (2/5)
5 Service Level Agreement (SLA) Introduction (3/5)
6 Introduction (4/5) many aspects of the I/O service are so dynamic that any static bounds cannot capture the real performance demands I/O requests can be very bursty requests in the burst period share the same latency bound with those in the quiet period? request size can be highly variable one common latency bound for small and large requests? spatial locality of requests can vary one common throughput bound for random and sequential requests? solution: Use Reference Storage System (RSS) as Performance Interface
7 Introduction (5/5)
8 Related Work (1/2) Research on Characterizing Storage Systems Performance there are 3 methods to characterize performance behaviors, namely, analytic device modeling, simulation, and black-box modeling Worthington et al. developed DiskSim, which represents the state of the art in disk simulation DiskSim models almost all performance-relevant components of a disk, including device drivers, buses, controllers, and caches
9 Related Work (2/2) compared with simulations, analytic models are much faster but cannot capture as many details as simulators black-box modeling, without assuming the knowledge of storage system s internal components or algorithms the training data set are fed into a machine learning model and then predict its response to an input the accuracy of the method relies on the selection of training data set and design of feature vectors
10 Proposed Scheme (1/5) Interpret RSS for the I/O scheduler to implement the interface use the classification and regression tree (CART) machine-learning model to predict known for its efficiency and accuracy predict what the latency of an arriving request is if it was received by RSS Efficiently implement the RSS interface meet simultaneously RSS requirements for different VSDs able to exploit request locality for system efficiency Migrate virtual storage devices (VSDs) for high device utilization different disk arrays exhibit various efficiency in hosting VSDs migrate VSDs to host arrays for high efficiency
11 feature vector: the measures includes request size, on-disk location of requested data, and the gap between the locations of two consecutive requests
12 Proposed Scheme (3/5) N+1 clocks N reference clocks (ref_clock) for N VSDs 1 wall clock (wall_clock) when the stream is considered for scheduling if its request is dispatched, then ref_clock += ref_service_time no pending requests, then ref_clock = wall_clock 09:00 09:20 09:40 10:20 09:20 10:00 09:05 09:15 09:25 09:10 09:13 09:16 09:22 09:19 09:28 10:00 10:00
13 Proposed Scheme (4/5) interference of multiple VSDs hosted on the same physical system can be a performance barrier YouChoose batches requests from the same VSD into time slots and schedules requests from the same time slot
14 Proposed Scheme (5/5) Migrating VSDs goal: to make the host system most efficiently used so as to accommodate as many VSDs as possible random workload sequential workload if currently VSD1 on HDD array A and VSD2 on SSD array, both arrays can significantly reduce their loads if we switch the hosts of the two VSDs VSD4 is insignificant in terms of resource consumption, and should be excluded from migration
15 Performance Evaluation (1/7) Disk arrays simulated by DiskSim QUANTUN_TORNADO with 10025RPMs and 1.245ms (fast disk) SEAGATE_ST32171W with 7200RPMs and 1.943ms (slow disk) Synthetic traces request size: 4 KB spatial locality from 0% to 100% the probability of two consecutive requests for contiguous data four Real-world I/O traces OpenMail: collected on a production system running the HP OpenMail WebSearch: collected from a popular search engine Financial: collected when an OLTP application runs in a financial institute VideoStreaming: collected when playing movies ( sequential reads)
16 Performance Evaluation (2/7) accuracy of the RSS interface the request s relative error, which is the ratio of the absolute error between the service time of each request predicted by the model and service time given by the RSS Fig(a) the average relative error for all the requests is 24% Fig(c) when the window size is 0.08s, more than 85% of the relative errors are smaller than 15%
17 Performance Evaluation (3/7) the accuracy of the model is acceptable with 80% of the errors less than 30% even if without sample workload
18 Performance Evaluation (4/7) impact of spatial locality Fig(a) on dedicated RSS Fig(b) on shared storage with YouChoose Fig(c) on shared storage using the 100 IOPS bound result 1: throughput is correlated to the workload s spatial locality result 2: YouChoose faithfully implements the performance interface specified with a chosen RSS, independent of spatial locality
19 Performance Evaluation (5/7) performance isolation Fig(a) on dedicated RSS Fig(b) on shared storage with YouChoose Fig(c) on shared storage using 100 IOPS bound workload 1: three segments of requests, each with a differing spatial locality (100%, 75%, and 0%), running on one VSD workload 2: 75% spatial locality and runs on another VSD throughput of workload is accordingly reduced whenever workload 1 at 165 th second, workload 2 ends, host can use reduces its spatial locality 100% 75% 0% the released resource to serve workload 1
20 Performance Evaluation (6/7) performance isolation Fig(a) real-world workloads on dedicated RSS Fig(b) real-world workloads workload 1: segments of WebSearch trace workload 2: segments of OpenMail, WebSearch, and Financial, one-after-one OpenMail WebSearch Financial
21 Performance Evaluation (7/7) random migration threshold = 10% * (1) 31.0% % = 16.5% > migration threshold = 10% (2) SSD Array spare capability = 100% % % = 71.3% (3) do migration, VSD7 is migrated to SSD Array sequentiality factor = 0.3 and 0.7, 24% and 20% more VSDs can be held in the system tradeoff between VSD migration and the improved efficiency
22 Conclusions Introduce Reference Storage System (RSS) as the performance interface Dynamic access behavior is well accommodated in the interface Use the machine learning technique to implement the RSS interface Achieve system efficiency with batched request scheduling
Spatial-Locality-Aware Virtual Storage Devices with Tangible QoS Expressions
Spatial-Locality-Aware Virtual Storage Devices with Tangible QoS Expressions Pei Yan, Song Jiang Dept. of Electrical & Computer Engineering Wayne State University Detroit, MI 4822, USA pyan@wayne.edu,
More informationNested QoS: Providing Flexible Performance in Shared IO Environment
Nested QoS: Providing Flexible Performance in Shared IO Environment Hui Wang Peter Varman Rice University Houston, TX 1 Outline Introduction System model Analysis Evaluation Conclusions and future work
More informationS-FTL: An Efficient Address Translation for Flash Memory by Exploiting Spatial Locality
S-FTL: An Efficient Address Translation for Flash Memory by Exploiting Spatial Locality Song Jiang, Lei Zhang, Xinhao Yuan, Hao Hu, and Yu Chen Department of Electrical and Computer Engineering Wayne State
More informationPerformance Modeling and Analysis of Flash based Storage Devices
Performance Modeling and Analysis of Flash based Storage Devices H. Howie Huang, Shan Li George Washington University Alex Szalay, Andreas Terzis Johns Hopkins University MSST 11 May 26, 2011 NAND Flash
More informationI/O Characterization of Commercial Workloads
I/O Characterization of Commercial Workloads Kimberly Keeton, Alistair Veitch, Doug Obal, and John Wilkes Storage Systems Program Hewlett-Packard Laboratories www.hpl.hp.com/research/itc/csl/ssp kkeeton@hpl.hp.com
More informationGetting it Right: Testing Storage Arrays The Way They ll be Used
Getting it Right: Testing Storage Arrays The Way They ll be Used Peter Murray Virtual Instruments Flash Memory Summit 2017 Santa Clara, CA 1 The Journey: How Did we Get Here? Storage testing was black
More informationQoS Support for End Users of I/O-intensive Applications using Shared Storage Systems
QoS Support for End Users of I/O-intensive Applications using Shared Storage Systems Xuechen Zhang ECE Department Wayne State University Detroit, MI, 4822, USA xczhang@wayne.edu Kei Davis Los Alamos National
More informationCharacterizing Data-Intensive Workloads on Modern Disk Arrays
Characterizing Data-Intensive Workloads on Modern Disk Arrays Guillermo Alvarez, Kimberly Keeton, Erik Riedel, and Mustafa Uysal Labs Storage Systems Program {galvarez,kkeeton,riedel,uysal}@hpl.hp.com
More informationDecoupling Datacenter Studies from Access to Large-Scale Applications: A Modeling Approach for Storage Workloads
Decoupling Datacenter Studies from Access to Large-Scale Applications: A Modeling Approach for Storage Workloads Christina Delimitrou 1, Sriram Sankar 2, Kushagra Vaid 2, Christos Kozyrakis 1 1 Stanford
More informationSamsung PM863 and SM863 for Data Centers
Samsung PM863 and SM863 for Data Centers Groundbreaking SSDs that raise the bar on satisfying bid data demands 2015 Samsung Electronics Co. An enormous increase in Internet traffic and data volume is challenging
More informationUsing Synology SSD Technology to Enhance System Performance Synology Inc.
Using Synology SSD Technology to Enhance System Performance Synology Inc. Synology_WP_ 20121112 Table of Contents Chapter 1: Enterprise Challenges and SSD Cache as Solution Enterprise Challenges... 3 SSD
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 informationEnergy Management Issue in Ad Hoc Networks
Wireless Ad Hoc and Sensor Networks - Energy Management Outline Energy Management Issue in ad hoc networks WS 2010/2011 Main Reasons for Energy Management in ad hoc networks Classification of Energy Management
More informationCondusiv s V-locity VM Accelerates Exchange 2010 over 60% on Virtual Machines without Additional Hardware
openbench Labs Executive Briefing: March 13, 2013 Condusiv s V-locity VM Accelerates Exchange 2010 over 60% on Virtual Machines without Additional Hardware Optimizing I/O for Increased Throughput and Reduced
More informationEnergy Management Issue in Ad Hoc Networks
Wireless Ad Hoc and Sensor Networks (Energy Management) Outline Energy Management Issue in ad hoc networks WS 2009/2010 Main Reasons for Energy Management in ad hoc networks Classification of Energy Management
More informationThe Pennsylvania State University. The Graduate School. Department of Computer Science and Engineering
The Pennsylvania State University The Graduate School Department of Computer Science and Engineering QUALITY OF SERVICE BENEFITS OF FLASH IN A STORAGE SYSTEM A Thesis in Computer Science and Engineering
More informationSolid State Drive (SSD) Cache:
Solid State Drive (SSD) Cache: Enhancing Storage System Performance Application Notes Version: 1.2 Abstract: This application note introduces Storageflex HA3969 s Solid State Drive (SSD) Cache technology
More informationDatabase Workload. from additional misses in this already memory-intensive databases? interference could be a problem) Key question:
Database Workload + Low throughput (0.8 IPC on an 8-wide superscalar. 1/4 of SPEC) + Naturally threaded (and widely used) application - Already high cache miss rates on a single-threaded machine (destructive
More informationGLUSTER CAN DO THAT! Architecting and Performance Tuning Efficient Gluster Storage Pools
GLUSTER CAN DO THAT! Architecting and Performance Tuning Efficient Gluster Storage Pools Dustin Black Senior Architect, Software-Defined Storage @dustinlblack 2017-05-02 Ben Turner Principal Quality Engineer
More informationParallel-DFTL: A Flash Translation Layer that Exploits Internal Parallelism in Solid State Drives
Parallel-: A Flash Translation Layer that Exploits Internal Parallelism in Solid State Drives Wei Xie, Yong Chen and Philip C. Roth Department of Computer Science, Texas Tech University, Lubbock, TX 7943
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 informationToward SLO Complying SSDs Through OPS Isolation
Toward SLO Complying SSDs Through OPS Isolation October 23, 2015 Hongik University UNIST (Ulsan National Institute of Science & Technology) Sam H. Noh 1 Outline Part 1: FAST 2015 Part 2: Beyond FAST 2
More informationWarming up Storage-level Caches with Bonfire
Warming up Storage-level Caches with Bonfire Yiying Zhang Gokul Soundararajan Mark W. Storer Lakshmi N. Bairavasundaram Sethuraman Subbiah Andrea C. Arpaci-Dusseau Remzi H. Arpaci-Dusseau 2 Does on-demand
More informationQuantifying Load Imbalance on Virtualized Enterprise Servers
Quantifying Load Imbalance on Virtualized Enterprise Servers Emmanuel Arzuaga and David Kaeli Department of Electrical and Computer Engineering Northeastern University Boston MA 1 Traditional Data Centers
More informationIntroduction Optimizing applications with SAO: IO characteristics Servers: Microsoft Exchange... 5 Databases: Oracle RAC...
HP StorageWorks P2000 G3 FC MSA Dual Controller Virtualization SAN Starter Kit Protecting Critical Applications with Server Application Optimization (SAO) Technical white paper Table of contents Introduction...
More informationApplication agnostic, empirical modelling of end-to-end memory hierarchy
Application agnostic, empirical modelling of end-to-end memory hierarchy Dhishankar Sengupta, Dell Inc. Krishanu Dhar, Dell Inc. Pradip Mukhopadhyay, NetApp Inc. Abstract Modeling(Analytical/trace-based/simulation)
More informationHyperthreading Technology
Hyperthreading Technology Aleksandar Milenkovic Electrical and Computer Engineering Department University of Alabama in Huntsville milenka@ece.uah.edu www.ece.uah.edu/~milenka/ Outline What is hyperthreading?
More informationFalcon: Scaling IO Performance in Multi-SSD Volumes. The George Washington University
Falcon: Scaling IO Performance in Multi-SSD Volumes Pradeep Kumar H Howie Huang The George Washington University SSDs in Big Data Applications Recent trends advocate using many SSDs for higher throughput
More informationPower Management in Storage Systems
Tag line, tag line Power Management in Storage Systems Kaladhar Voruganti Technical Director CTO Office, Sunnyvale June 12, 2009 Outline Power Consumption Background in Data Centers and Storage Systems
More informationPARDA: Proportional Allocation of Resources for Distributed Storage Access
PARDA: Proportional Allocation of Resources for Distributed Storage Access Ajay Gulati, Irfan Ahmad, Carl Waldspurger Resource Management Team VMware Inc. USENIX FAST 09 Conference February 26, 2009 The
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 informationTechnical Note. Improving Random Read Performance Using Micron's SNAP READ Operation. Introduction. TN-2993: SNAP READ Operation.
Introduction Technical Note Improving Random Read Performance Using Micron's SNAP READ Operation Introduction NAND flash devices are designed for applications that require nonvolatile, high-density, fast
More informationHP EVA P6000 Storage performance
Technical white paper HP EVA P6000 Storage performance Table of contents Introduction 2 Sizing up performance numbers 2 End-to-end performance numbers 3 Cache performance numbers 4 Performance summary
More informationMicrosoft Exchange Server 2010 workload optimization on the new IBM PureFlex System
Microsoft Exchange Server 2010 workload optimization on the new IBM PureFlex System Best practices Roland Mueller IBM Systems and Technology Group ISV Enablement April 2012 Copyright IBM Corporation, 2012
More informationNested QoS: Providing Flexible Performance in Shared IO Environment
Nested QoS: Providing Flexible Performance in Shared IO Environment Hui Wang Peter Varman hw5@rice.edu pjv@rice.edu Rice University, USA Abstract The increasing popularity of storage and server consolidation
More informationUnderstanding the Relation between the Performance and Reliability of NAND Flash/SCM Hybrid Solid- State Drive
Understanding the Relation between the Performance and Reliability of NAND Flash/SCM Hybrid Solid- State Drive Abstract: A NAND flash memory/storage-class memory (SCM) hybrid solid-state drive (SSD) can
More informationBalancing Fairness and Efficiency in Tiered Storage Systems with Bottleneck-Aware Allocation
Balancing Fairness and Efficiency in Tiered Storage Systems with Bottleneck-Aware Allocation Hui Wang, Peter Varman Rice University FAST 14, Feb 2014 Tiered Storage Tiered storage: HDs and SSDs q Advantages:
More informationNimble Storage Adaptive Flash
Nimble Storage Adaptive Flash Read more Nimble solutions Contact Us 800-544-8877 solutions@microage.com MicroAge.com TECHNOLOGY OVERVIEW Nimble Storage Adaptive Flash Nimble Storage s Adaptive Flash platform
More informationI/O Commercial Workloads. Scalable Disk Arrays. Scalable ICDA Performance. Outline of This Talk: Related Work on Disk Arrays.
Scalable Disk Arrays I/O Commercial Workloads David Kaeli Northeastern University Computer Architecture Research Laboratory Boston, MA Manpreet Singh William Zahavi EMC Corporation Hopkington, MA Industry
More informationTechnical Paper. Performance and Tuning Considerations for SAS on Dell EMC VMAX 250 All-Flash Array
Technical Paper Performance and Tuning Considerations for SAS on Dell EMC VMAX 250 All-Flash Array Release Information Content Version: 1.0 April 2018 Trademarks and Patents SAS Institute Inc., SAS Campus
More informationA Comparative Study of Microsoft Exchange 2010 on Dell PowerEdge R720xd with Exchange 2007 on Dell PowerEdge R510
A Comparative Study of Microsoft Exchange 2010 on Dell PowerEdge R720xd with Exchange 2007 on Dell PowerEdge R510 Incentives for migrating to Exchange 2010 on Dell PowerEdge R720xd Global Solutions Engineering
More informationA Novel Buffer Management Scheme for SSD
A Novel Buffer Management Scheme for SSD Qingsong Wei Data Storage Institute, A-STAR Singapore WEI_Qingsong@dsi.a-star.edu.sg Bozhao Gong National University of Singapore Singapore bzgong@nus.edu.sg Cheng
More informationCS 426 Parallel Computing. Parallel Computing Platforms
CS 426 Parallel Computing Parallel Computing Platforms Ozcan Ozturk http://www.cs.bilkent.edu.tr/~ozturk/cs426/ Slides are adapted from ``Introduction to Parallel Computing'' Topic Overview Implicit Parallelism:
More informationEvaluation Report: Improving SQL Server Database Performance with Dot Hill AssuredSAN 4824 Flash Upgrades
Evaluation Report: Improving SQL Server Database Performance with Dot Hill AssuredSAN 4824 Flash Upgrades Evaluation report prepared under contract with Dot Hill August 2015 Executive Summary Solid state
More informationSamsung SSD PM863 and SM863 for Data Centers. Groundbreaking SSDs that raise the bar on satisfying big data demands
Samsung SSD PM863 and SM863 for Data Centers Groundbreaking SSDs that raise the bar on satisfying big data demands 2 Samsung SSD PM863 and SM863 Innovations in solid state As the importance of data in
More informationPerformance Analysis in the Real World of Online Services
Performance Analysis in the Real World of Online Services Dileep Bhandarkar, Ph. D. Distinguished Engineer 2009 IEEE International Symposium on Performance Analysis of Systems and Software My Background:
More informationCBM: A Cooperative Buffer Management for SSD
3 th International Conference on Massive Storage Systems and Technology (MSST 4) : A Cooperative Buffer Management for SSD Qingsong Wei, Cheng Chen, Jun Yang Data Storage Institute, A-STAR, Singapore June
More informationDongjun Shin Samsung Electronics
2014.10.31. Dongjun Shin Samsung Electronics Contents 2 Background Understanding CPU behavior Experiments Improvement idea Revisiting Linux I/O stack Conclusion Background Definition 3 CPU bound A computer
More informationHibachi: A Cooperative Hybrid Cache with NVRAM and DRAM for Storage Arrays
Hibachi: A Cooperative Hybrid Cache with NVRAM and DRAM for Storage Arrays Ziqi Fan, Fenggang Wu, Dongchul Park 1, Jim Diehl, Doug Voigt 2, and David H.C. Du University of Minnesota, 1 Intel, 2 HP Enterprise
More informationDatabase Architecture 2 & Storage. Instructor: Matei Zaharia cs245.stanford.edu
Database Architecture 2 & Storage Instructor: Matei Zaharia cs245.stanford.edu Summary from Last Time System R mostly matched the architecture of a modern RDBMS» SQL» Many storage & access methods» Cost-based
More informationWHITE PAPER. Optimizing Virtual Platform Disk Performance
WHITE PAPER Optimizing Virtual Platform Disk Performance Optimizing Virtual Platform Disk Performance 1 The intensified demand for IT network efficiency and lower operating costs has been driving the phenomenal
More informationModule 1: Basics and Background Lecture 4: Memory and Disk Accesses. The Lecture Contains: Memory organisation. Memory hierarchy. Disks.
The Lecture Contains: Memory organisation Example of memory hierarchy Memory hierarchy Disks Disk access Disk capacity Disk access time Typical disk parameters Access times file:///c /Documents%20and%20Settings/iitkrana1/My%20Documents/Google%20Talk%20Received%20Files/ist_data/lecture4/4_1.htm[6/14/2012
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 informationRAID4S: Improving RAID Performance with Solid State Drives
RAID4S: Improving RAID Performance with Solid State Drives Rosie Wacha UCSC: Scott Brandt and Carlos Maltzahn LANL: John Bent, James Nunez, and Meghan Wingate SRL/ISSDM Symposium October 19, 2010 1 RAID:
More informationPowerVault MD3 SSD Cache Overview
PowerVault MD3 SSD Cache Overview A Dell Technical White Paper Dell Storage Engineering October 2015 A Dell Technical White Paper TECHNICAL INACCURACIES. THE CONTENT IS PROVIDED AS IS, WITHOUT EXPRESS
More informationCopyright 2012 EMC Corporation. All rights reserved.
1 FLASH 1 ST THE STORAGE STRATEGY FOR THE NEXT DECADE Richard Gordon EMEA FLASH Business Development 2 Information Tipping Point Ahead The Future Will Be Nothing Like The Past 140,000 120,000 100,000 80,000
More informationMigration Based Page Caching Algorithm for a Hybrid Main Memory of DRAM and PRAM
Migration Based Page Caching Algorithm for a Hybrid Main Memory of DRAM and PRAM Hyunchul Seok Daejeon, Korea hcseok@core.kaist.ac.kr Youngwoo Park Daejeon, Korea ywpark@core.kaist.ac.kr Kyu Ho Park Deajeon,
More informationdavidklee.net gplus.to/kleegeek linked.com/a/davidaklee
@kleegeek davidklee.net gplus.to/kleegeek linked.com/a/davidaklee Specialties / Focus Areas / Passions: Performance Tuning & Troubleshooting Virtualization Cloud Enablement Infrastructure Architecture
More informationTowards Model-based Management of Database Fragmentation
Towards Model-based Management of Database Fragmentation Asim Ali, Abdelkarim Erradi, Rashid Hadjidj Qatar University Rui Jia, Sherif Abdelwahed Mississippi State University Outline p Introduction p Model-based
More informationForget IOPS: A Proper Way to Characterize & Test Storage Performance Peter Murray SwiftTest
Forget IOPS: A Proper Way to Characterize & Test Storage Performance Peter Murray peter@swifttest.com SwiftTest Storage Performance Validation Rely on vendor IOPS claims Test in production and pray Validate
More informationSHARDS & Talus: Online MRC estimation and optimization for very large caches
SHARDS & Talus: Online MRC estimation and optimization for very large caches Nohhyun Park CloudPhysics, Inc. Introduction Efficient MRC Construction with SHARDS FAST 15 Waldspurger at al. Talus: A simple
More information15-740/ Computer Architecture Lecture 20: Main Memory II. Prof. Onur Mutlu Carnegie Mellon University
15-740/18-740 Computer Architecture Lecture 20: Main Memory II Prof. Onur Mutlu Carnegie Mellon University Today SRAM vs. DRAM Interleaving/Banking DRAM Microarchitecture Memory controller Memory buses
More informationHybrid Storage Architecture Marries Performance and Efficiency
Hybrid Storage Architecture Marries Performance and Efficiency Bill Mottram VP of Marketing at Atrato Au to nom ic 2010 Storage [aw-tuh-nom-ik Developer Conference. ] Insert Your Company Name. All Rights
More informationNo Tradeoff Low Latency + High Efficiency
No Tradeoff Low Latency + High Efficiency Christos Kozyrakis http://mast.stanford.edu Latency-critical Applications A growing class of online workloads Search, social networking, software-as-service (SaaS),
More informationArchitectural Differences nc. DRAM devices are accessed with a multiplexed address scheme. Each unit of data is accessed by first selecting its row ad
nc. Application Note AN1801 Rev. 0.2, 11/2003 Performance Differences between MPC8240 and the Tsi106 Host Bridge Top Changwatchai Roy Jenevein risc10@email.sps.mot.com CPD Applications This paper discusses
More informationarxiv: v1 [cs.ar] 11 Apr 2017
FMMU: A Hardware-Automated Flash Map Management Unit for Scalable Performance of NAND Flash-Based SSDs Yeong-Jae Woo Sang Lyul Min Department of Computer Science and Engineering, Seoul National University
More informationBuilding Adaptive Performance Models for Dynamic Resource Allocation in Cloud Data Centers
Building Adaptive Performance Models for Dynamic Resource Allocation in Cloud Data Centers Jin Chen University of Toronto Joint work with Gokul Soundararajan and Prof. Cristiana Amza. Today s Cloud Pay
More informationI/O Buffering and Streaming
I/O Buffering and Streaming I/O Buffering and Caching I/O accesses are reads or writes (e.g., to files) Application access is arbitary (offset, len) Convert accesses to read/write of fixed-size blocks
More informationDeep Learning Performance and Cost Evaluation
Micron 5210 ION Quad-Level Cell (QLC) SSDs vs 7200 RPM HDDs in Centralized NAS Storage Repositories A Technical White Paper Don Wang, Rene Meyer, Ph.D. info@ AMAX Corporation Publish date: October 25,
More informationPreface. Fig. 1 Solid-State-Drive block diagram
Preface Solid-State-Drives (SSDs) gained a lot of popularity in the recent few years; compared to traditional HDDs, SSDs exhibit higher speed and reduced power, thus satisfying the tough needs of mobile
More informationBest Practices for Setting BIOS Parameters for Performance
White Paper Best Practices for Setting BIOS Parameters for Performance Cisco UCS E5-based M3 Servers May 2013 2014 Cisco and/or its affiliates. All rights reserved. This document is Cisco Public. Page
More informationA Predictable RTOS. Mantis Cheng Department of Computer Science University of Victoria
A Predictable RTOS Mantis Cheng Department of Computer Science University of Victoria Outline I. Analysis of Timeliness Requirements II. Analysis of IO Requirements III. Time in Scheduling IV. IO in Scheduling
More informationIBM DS8870 Release 7.0 Performance Update
IBM DS8870 Release 7.0 Performance Update Enterprise Storage Performance David Whitworth Yan Xu 2012 IBM Corporation Agenda Performance Overview System z (CKD) Open Systems (FB) Easy Tier Copy Services
More informationPRESERVE DATABASE PERFORMANCE WHEN RUNNING MIXED WORKLOADS
PRESERVE DATABASE PERFORMANCE WHEN RUNNING MIXED WORKLOADS Testing shows that a Pure Storage FlashArray//m storage array used for Microsoft SQL Server 2016 helps eliminate latency and preserve productivity.
More informationDelayed Partial Parity Scheme for Reliable and High-Performance Flash Memory SSD
Delayed Partial Parity Scheme for Reliable and High-Performance Flash Memory SSD Soojun Im School of ICE Sungkyunkwan University Suwon, Korea Email: lang33@skku.edu Dongkun Shin School of ICE Sungkyunkwan
More informationAPP: MINIMIZING INTERFERENCE USING AGGRESSIVEPIPELINED PREFETCHING IN MULTI-LEVEL BUFFER CACHES
APP: MINIMIZING INTERFERENCE USING AGGRESSIVEPIPELINED PREFETCHING IN MULTI-LEVEL BUFFER CACHES Christina M. Patrick Pennsylvania State University patrick@cse.psu.edu Nicholas Voshell Pennsylvania State
More informationSmartSaver: Turning Flash Drive into a Disk Energy Saver for Mobile Computers
SmartSaver: Turning Flash Drive into a Disk Energy Saver for Mobile Computers Feng Chen 1 Song Jiang 2 Xiaodong Zhang 1 The Ohio State University, USA Wayne State University, USA Disks Cost High Energy
More informationNovel Address Mappings for Shingled Write Disks
Novel Address Mappings for Shingled Write Disks Weiping He and David H.C. Du Department of Computer Science, University of Minnesota, Twin Cities {weihe,du}@cs.umn.edu Band Band Band Abstract Shingled
More informationIsilon Performance. Name
1 Isilon Performance Name 2 Agenda Architecture Overview Next Generation Hardware Performance Caching Performance Streaming Reads Performance Tuning OneFS Architecture Overview Copyright 2014 EMC Corporation.
More informationA Better Storage Solution
A Better Storage Solution Presented by: Richard Goss Presentation to The Problem with Hard Disks Processors have increased in speed by orders of magnitude over the years. But spinning hard disk drives
More informationCross-layer Optimization for Virtual Machine Resource Management
Cross-layer Optimization for Virtual Machine Resource Management Ming Zhao, Arizona State University Lixi Wang, Amazon Yun Lv, Beihang Universituy Jing Xu, Google http://visa.lab.asu.edu Virtualized Infrastructures,
More informationStorage Workload Isolation via Tier Warming: How Models Can Help
Storage Workload Isolation via Tier Warming: How Models Can Help By Ji Xue, Feng Yan, Alma Riska, and Evgenia Smirni Presented By Christian Contreras Proposed solution Storage workload prediction model
More informationAzor: Using Two-level Block Selection to Improve SSD-based I/O caches
Azor: Using Two-level Block Selection to Improve SSD-based I/O caches Yannis Klonatos, Thanos Makatos, Manolis Marazakis, Michail D. Flouris, Angelos Bilas {klonatos, makatos, maraz, flouris, bilas}@ics.forth.gr
More informationAssert(!Defined(Sequential I/O)) Cheng Li*, Philip Shilane, Fred Douglis, Darren Sawyer, and Hyong Shim
Assert(!Defined(Sequential I/O)) Cheng Li*, Philip Shilane, Fred Douglis, Darren Sawyer, and Hyong Shim *utgers University EMC Corporation 1 Sequential I/O is Important Driven by traditional storage characteristics
More informationNimble Storage vs HPE 3PAR: A Comparison Snapshot
Nimble Storage vs HPE 3PAR: A 1056 Baker Road Dexter, MI 48130 t. 734.408.1993 Nimble Storage vs HPE 3PAR: A INTRODUCTION: Founders incorporated Nimble Storage in 2008 with a mission to provide customers
More informationUC Santa Cruz UC Santa Cruz Electronic Theses and Dissertations
UC Santa Cruz UC Santa Cruz Electronic Theses and Dissertations Title Providing Predictable Performance in Flash and Black-box Storage Permalink https://escholarship.org/uc/item/4wj9r8np Author Skourtis,
More informationIssues and Challenges in the Performance Analysis of Real Disk Arrays
IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, VOL. 15, NO. 6, JUNE 2004 559 Issues and Challenges in the Performance Analysis of Real Disk Arrays Elizabeth Varki, Member, IEEE, Arif Merchant,
More informationFinal Lecture. A few minutes to wrap up and add some perspective
Final Lecture A few minutes to wrap up and add some perspective 1 2 Instant replay The quarter was split into roughly three parts and a coda. The 1st part covered instruction set architectures the connection
More informationSTLAC: A Spatial and Temporal Locality-Aware Cache and Networkon-Chip
STLAC: A Spatial and Temporal Locality-Aware Cache and Networkon-Chip Codesign for Tiled Manycore Systems Mingyu Wang and Zhaolin Li Institute of Microelectronics, Tsinghua University, Beijing 100084,
More informationMeasuring VDI Fitness and User Experience Technical White Paper
Measuring VDI Fitness and User Experience Technical White Paper 3600 Mansell Road Suite 200 Alpharetta, GA 30022 866.914.9665 main 678.397.0339 fax info@liquidwarelabs.com www.liquidwarelabs.com Table
More informationTechnical Paper. Performance and Tuning Considerations for SAS on the Hitachi Virtual Storage Platform G1500 All-Flash Array
Technical Paper Performance and Tuning Considerations for SAS on the Hitachi Virtual Storage Platform G1500 All-Flash Array Release Information Content Version: 1.0 April 2018. Trademarks and Patents SAS
More informationLSM-trie: An LSM-tree-based Ultra-Large Key-Value Store for Small Data
LSM-trie: An LSM-tree-based Ultra-Large Key-Value Store for Small Data Xingbo Wu Yuehai Xu Song Jiang Zili Shao The Hong Kong Polytechnic University The Challenge on Today s Key-Value Store Trends on workloads
More informationVirtual Storage Tier and Beyond
Virtual Storage Tier and Beyond Manish Agarwal Sr. Product Manager, NetApp Santa Clara, CA 1 Agenda Trends Other Storage Trends and Flash 5 Min Rule Issues for Flash Dedupe and Flash Caching Architectural
More informationA Scheduling Framework that Makes any Disk Schedulers Non-work-conserving solely based on Request Characteristics
A Scheduling Framework that Makes any Disk Schedulers Non-work-conserving solely based on Request Characteristics Yuehai Xu ECE Department Wayne State University Detroit, MI 48202, USA yhxu@wayne.edu Song
More informationNEC Express5800 A2040b 22TB Data Warehouse Fast Track. Reference Architecture with SW mirrored HGST FlashMAX III
NEC Express5800 A2040b 22TB Data Warehouse Fast Track Reference Architecture with SW mirrored HGST FlashMAX III Based on Microsoft SQL Server 2014 Data Warehouse Fast Track (DWFT) Reference Architecture
More informationCS 425 / ECE 428 Distributed Systems Fall 2015
CS 425 / ECE 428 Distributed Systems Fall 2015 Indranil Gupta (Indy) Measurement Studies Lecture 23 Nov 10, 2015 Reading: See links on website All Slides IG 1 Motivation We design algorithms, implement
More informationCopyright 2012, Oracle and/or its affiliates. All rights reserved.
1 Storage Innovation at the Core of the Enterprise Robert Klusman Sr. Director Storage North America 2 The following is intended to outline our general product direction. It is intended for information
More informationNext-Generation Cloud Platform
Next-Generation Cloud Platform Jangwoo Kim Jun 24, 2013 E-mail: jangwoo@postech.ac.kr High Performance Computing Lab Department of Computer Science & Engineering Pohang University of Science and Technology
More informationAll-Flash Storage Solution for SAP HANA:
All-Flash Storage Solution for SAP HANA: Storage Considerations using SanDisk Solid State Devices WHITE PAPER Western Digital Technologies, Inc. 951 SanDisk Drive, Milpitas, CA 95035 www.sandisk.com Table
More informationFlash Trends: Challenges and Future
Flash Trends: Challenges and Future John D. Davis work done at Microsoft Researcher- Silicon Valley in collaboration with Laura Caulfield*, Steve Swanson*, UCSD* 1 My Research Areas of Interest Flash characteristics
More information