Moving to SANless for your Database Storage
|
|
- Rosanna Manning
- 6 years ago
- Views:
Transcription
1 Moving to SANless for your Database Storage Back to DASD Tony Rogerson, SQL Server
2
3 Transition Back (SAN to DASD) Node1 Node2 Node1 Node2 Shared Storage DASD DASD Disk Replication Shared Storage Node2 DASD
4 Agenda Kit Connectivity Protocols Mechanical Drives Solid State Storage SANless SQL Server Semi-Random GUID V Incremental ID Comparison Clustered V Heap Bottleneck shift (CPU, Latching) TempDB Transaction Logs Scale out approach Availability Groups Hybrid Tiered Storage Approach Futures Massively Parallel Computing (GPGPU)
5 Getting to your device CONNECTIVITY PROTOCOLS
6 Connectivity Protocols Cable attached SAS and SATA 3Gb/s (300MiB/s) SAS and SATA 6Gb/s (600MiB/s) SATA (3/6Gbit/s) InfiniBand (GiBytes/s) Fibre Channel (GiBytes/s) External PCIe (GiBytes/s) USB 2.0, 3.0 (ok, MiBytes/s but latency shocking) Directly Attached PCIe (GiBytes/s)
7 Point to Point (SAS/SATA) SAS 600 / SATA 3 550MiBytes/sec SAS 600 / SATA 3 550MiBytes/sec SAS 600 / SATA 3 550MiBytes/sec PCIe SAS 600 / SATA 3 550MiBytes/sec Each channel has full throughput (not shared)
8 Port Expanders/Multipliers SAS 600 / SATA 3 550MiBytes/sec (Multiplexed) PCIe Devices share the bandwidth (same as Ultra SCSI multi-drop bus architecture). OK for HDD, not for SSD
9 Throughput Limiters 10 x SAS 600 / SATA 3 10 x 550MiBytes/sec SSD populated SAN 4Gbit/sec HBA SERVER Bottleneck is HBA, multiple HBA s required more cost! SSD easily run at 500MiB/sec on sequential and random loads. 4Gbit = approx 500MiB/sec 5.5GiBytes/s into 500MiB/s does not go
10 Rotating disk/spindle/hard disk whatever it has moving parts! MECHANICAL DRIVES
11 Mechanical Drives Stuck at 15Krpm for over a decade Density not improving (150GB per platter for enterprise 15Krpm, 1TB per platter for commodity 7,2Krpm) though helium filled coming Performance (latency and throughput) variance - outer edge better than inner. Poor level of IOps for a semi-random database workload. Ok for sequential. Power and heat high compared to Flash
12 Disk Throughput 100% Sequential 100% Random Strip Size: 64KiB; Windows OS allocation 8KiB; Transfer Size 8KiB 40GiB test file created on freshly formatted drive 146GB partition Many worker threads required to get throughput Drive is a HP EH0146FARWD 146GB K
13 Mechanics of a Hard Drive Multiple Platters per disk Platters double sided Two heads per platter Move together More sectors on outer than inner
14 Disk Geometry Effects 100% sequential read 64KiB transfer 3223IOps 0.30ms latency 1914IOps 0.52ms latency
15 Disk Geometry Visualisation Drop of 55,000 Kbytes per second from outer to inner edge Faster decline in performance as you approach the centre * HDDScan utility used
16 Ringing out the Performance (Mechanical drives) Increase platters [2 heads per platter], max 4 platters on K, 2 on K Hybrid add flash and keep hot data on the SSD Shortstroking RAID (striping a form of shortstroking!) Disk writes are smoothed through Write Cache (have plenty of it!)
17 Shortstroke the disk 3223IOps 0.30ms latency 1914IOps 0.52ms latency
18 Shortstroking Use only the sweet spot (outer edge) More sectors can be read in a single revolution Less head movement (jump tracks). 600GiB disk may only use 50GiB!
19 Effect of RAID MiBytes per second Latency (ms) # disks # disks More disks, less data per disk (implicit short stroking) Sequential workload saturates at 330MiB/sec regardless of number of disks port multiplexing? TEST TEST TEST your hardware before installing anything!
20 Master File Table Position FREE SPACE!!! Files will start to the right of the MFT so you end up with the sweet spot of the disk being free space 3.1GiB not used.
21 Like me when hung-over, it has no moving parts SOLID STATE STORAGE
22 Solid State Storage Game changer because of single device capability (IOps [random and sequential], Throughput, Latency, Power and heat. Commodity MLC, TLC Enterprise emlc, SLC, TLC, MLC Future? Density of flash is finite because of leakage between cells. SLC Single Level Cell: 1 bit per cell MLC Multi-Level Cell: 2 bits per cell TLC Triple Level Cell: 3 bits per cell
23 SSD Geometry (SQL Server Serial Plan) Consistent Throughput and Latency Query time 8mins 37secs Notes: Clustered Index on table; Table Scan to calculate MAX( CHECKSUM() ) on a char(7900) column;
24 SSD Geometry Effects (SQL Server Parallelism) Consistent Throughput and Latency Query time 3mins 7secs Notes: Clustered Index on table; Table Scan to calculate MAX( CHECKSUM() ) on a char(7900) column;
25 Disk Geometry Effects (SQL Server Parallelism) Latency increase gives Throughput decrease Query time 2hrs 50mins Notes: Clustered Index on table; Table Scan to calculate MAX( CHECKSUM() ) on a char(7900) column;
26 Disk Geometry Effects (SQL Server Serial Plan) Latency increase gives Throughput decrease Query time 2hrs 50mins Notes: Clustered Index on table; Table Scan to calculate MAX( CHECKSUM() ) on a char(7900) column;
27 NOLOCK effect When Nolock used it follows the page chain of the leaf level of the clustered index. Otherwise, it follows the B-Tree.
28 Performance P/E Cycle Cells cannot be overwritten Update = Erase then write Background processes perform erase Write goes to another cell (unlike a Hard Disk) TRIM works for the file system not internally for SQL Files which remain constant.
29 Reliability? Wear Levelling through Over Provisioning Smart algorithms Errors mostly occur on Write RAISE (redundant array independent silicon elements) Long endurance ratings (e.g. Intel 710) 1.8PB to 3PB for 300GiB drive
30 Form Factors SAS or SATA device 3.5 or 2.5 form factor Limited to protocol throughput (max 550MiB/sec) per channel Sizes up-to 2TB per drive PCIe Full or half height Limited to bus lane speed and capacity (16 lanes PCIe 2.0 gives approx 6.7GBytes/sec) Servers: each CPU socket has own PCI bus for performance more lanes, more throughput! Sizes up-to 10.24TB per card
31 Requirement: Cost Comparison I need to store 40GiBytes of data in a SQL Server table, the query I want to perform needs to return quickly, is highly random in terms of reading, anticipated requirement of 4,000 64KByte IOps at a read latency of less than 4 milliseconds per IO. Memory is limited to 8GiBytes.
32 6 x 15Krpm drives RAID 0 Cost Sequential Workload Single OCZ IBIS (PCIe) Random Workload Price per IO 304 * 6 / 4093 = 0.44 per IO Price per GB 304 * 6 / 40 = per GB Price per IO 442 / = per IO Price per GB 442 / 40 = per GB Requirement: Data is 40GiBytes, requirement is 4000 IOps for a single worker on a highly random query.
33 Full circle, but they re calling it Appliance SANLESS
34 Trend: [move Process TO data] not [move data TO process] Programs operate close to the data thus reducing latency and throughput bottlenecks Latency! Latency! Latency! (and throughput) Results PROCESS SAN (Network) DATA Results PCIe PROCESS DATA
35 What is a SAN? Storage Server connected via a Storage Area Network. Servers connected via the Storage Area Network using one or more HBA, multipathing for redundancy etc.. Storage Server OS is specifically tailored to the job of Storage.
36 Why SANs? Consolidate storage in one place solves the proliferation of DASD hard disks {jobs for the boys proprietary} Management easier provisioning of extra storage {nothing wrong with that!} Performance benefits IOps and throughput {now cheaper alternatives + I question the cost to get good IOps with low latency} SQL Server Clustering and the write Cache problem
37 Appliances? The big virtual appliance the cloud doesn t use them! Appliances all based on DASD PDW/Fast Track based around SAN but device dedicated to a single server node approach. Two prongs: Server based appliances with DASD replace SAN SAN based appliances (with SQL Server running inside) replace servers?? Does that not become DASD?
38 HA/DR? No more central point of failure the SAN Scale out multiple copies of the data kept in sync. SQL Server Availability Groups CITRIX NetScaler DataStream Other Scale out products HADOOP Cassandra VoltDB Gizzzzziiiiillliannnnn more.
39 Shared Storage or Scale out DASD Node1 Node2 Node1 Node2 Shared Storage DASD DASD Node2 Disk Replication Shared Storage DASD
40 Some considerations/thoughts SQL SERVER
41 SQL Server yes that old war GUID V INCREMENTAL-ID
42 The GUID war SSD s are fantastic at high IOp random IO at a low latency Inserts into HEAPS are blisteringly quick Do we care about fragmentation? For IO (sort of): Fragmentation = more IO because data isn t contiguous More memory usage More latching No right answer depends what you are doing!
43 SQL Server another religion to rethink COMPARISON CLUSTERED V HEAP
44 Data Benefits of clustering Leaf level of the index is the data Data is grouped in terms of locality by the clustering key (subject to extent fragmentation) No RID look up required when using clustering key
45 Data Benefits of Heap No B-tree maintenance causing page splits on un-ordered inserting Non-Clustered indexes narrow and smaller in size Insert speed rocks
46 Sample Lookup Data 50K rows randomly picked (using ORDER BY NEWID()) Fairly even pick of rows across the big table of 63,419,340 rows
47 Comparison DEMO
48 SQL Server BOTTLENECK SHIFT (CPU, LATCHING)
49 New bottleneck Driving the CPU harder data arrives earlier Exposes weakness in the application hidden by slow IO Instead of wait for IO now more latching in memory Helps reduce deadlocks because transactions are faster
50 SQL Server TEMPDB
51 TEMPDB on SSD? TEMPDB usage Version Store Work Tables for Group By, Order By etc.. User temporary tables and table variables How big is your tempdb requirement? Doable on SSD but provision based on expected throughput rate and expected ware rate
52 SQL Server TRANSACTION LOGS
53 Any point? Sequential writes to the outer portion of a disk is good already 12K IOP s at 0.08ms latency 96MiB/sec What s your transaction throughput rate? Can you really keep the transaction log file on the outer edge of the disk and the head in the correct track? Anyway Write cache should soak that up and smooth out
54 Reducing Transaction Log Perf on Mechanical Drives Single RAID array stick all the transaction logs from all the databases on it Disk head movement! Adding Latency Balanced by write cache? TEST TEST TEST
55 SQL Server SCALE OUT APPROACH (AVAILABILITY GROUPS)
56 Scale out HA/DR approach SERV_A DASD SERV_C Disaster Recovery DASD SERV_D DASD SERV_B DASD Cluster (Availability Group Listener) Readers (ReadOnly intent) (via Listener) Writers (via Listener) Readers
57 SQL Server HYBRID TIERED STORAGE
58 Scale Up (single box) Tier 0 File Group Tier 1 File Group Tier 2 File Group Tier 3 File Group PCIe Enterprise PCIe Commodity SAS Drives SATA Drives Table Placement Partitioning of current to historical data across file groups
59 From the few to the many GPGPU (MASSIVELY PARALLEL COMPUTING)
60 GPGPU v CPU comparison 192 logical proc s (GPGPU) v 8 (CPU) Host to Device (GPU) GiBytes/s depending on lanes + PCIe 2.0 or 3.0 Internal GPU memory speed astonishing
61 Summary Mechanical drives do they still have a place depends what you are doing High IOps use PCIe based cards Scale Out rather than Up Be fooled no longer by marketing BS Be aware what protocol version you are using PCIe 2.0 v 3.0, SATA 3 v 6, SAS 3 v 6. Finally: cut away all those components and go scale-out, design from the offset.
62 Remember: It s lunchtime, its sunny and you are now being held up. Tony Rogerson, SQL Server tonyrogerson@torver.net
63
64
SSD Architecture Considerations for a Spectrum of Enterprise Applications. Alan Fitzgerald, VP and CTO SMART Modular Technologies
SSD Architecture Considerations for a Spectrum of Enterprise Applications Alan Fitzgerald, VP and CTO SMART Modular Technologies Introduction Today s SSD delivers form-fit-function compatible solid-state
More informationInterface Trends for the Enterprise I/O Highway
Interface Trends for the Enterprise I/O Highway Mitchell Abbey Product Line Manager Enterprise SSD August 2012 1 Enterprise SSD Market Update One Size Does Not Fit All : Storage solutions will be tiered
More informationCOMP283-Lecture 3 Applied Database Management
COMP283-Lecture 3 Applied Database Management Introduction DB Design Continued Disk Sizing Disk Types & Controllers DB Capacity 1 COMP283-Lecture 3 DB Storage: Linear Growth Disk space requirements increases
More informationConsidering the 2.5-inch SSD-based RAID Solution:
Considering the 2.5-inch SSD-based RAID Solution: Using Infortrend EonStor B12 Series with Intel SSD in a Microsoft SQL Server Environment Application Note Abstract This application note discusses the
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 informationVirtual Memory. Reading. Sections 5.4, 5.5, 5.6, 5.8, 5.10 (2) Lecture notes from MKP and S. Yalamanchili
Virtual Memory Lecture notes from MKP and S. Yalamanchili Sections 5.4, 5.5, 5.6, 5.8, 5.10 Reading (2) 1 The Memory Hierarchy ALU registers Cache Memory Memory Memory Managed by the compiler Memory Managed
More informationStorage Optimization with Oracle Database 11g
Storage Optimization with Oracle Database 11g Terabytes of Data Reduce Storage Costs by Factor of 10x Data Growth Continues to Outpace Budget Growth Rate of Database Growth 1000 800 600 400 200 1998 2000
More informationIt Takes Guts to be Great
It Takes Guts to be Great Sean Stead, STEC Tutorial C-11: Enterprise SSDs Tues Aug 21, 2012 8:30 to 11:20AM 1 Who s Inside Your SSD? Full Data Path Protection Host Interface It s What s On The Inside That
More informationThe Oracle Database Appliance I/O and Performance Architecture
Simple Reliable Affordable The Oracle Database Appliance I/O and Performance Architecture Tammy Bednar, Sr. Principal Product Manager, ODA 1 Copyright 2012, Oracle and/or its affiliates. All rights reserved.
More informationPC-based data acquisition II
FYS3240 PC-based instrumentation and microcontrollers PC-based data acquisition II Data streaming to a storage device Spring 2015 Lecture 9 Bekkeng, 29.1.2015 Data streaming Data written to or read from
More informationZD-XL SQL Accelerator 1.6
ZD-XL SQL Accelerator 1.6 Integrated Flash Hardware & Software Acceleration Solution for SQL Server Features Supports Microsoft Hyper-V and VMware ESXi environments Accelerates SQL Server at a per database
More informationIntroduction to I/O and Disk Management
1 Secondary Storage Management Disks just like memory, only different Introduction to I/O and Disk Management Why have disks? Ø Memory is small. Disks are large. Short term storage for memory contents
More informationNVMe: The Protocol for Future SSDs
When do you need NVMe? You might have heard that Non-Volatile Memory Express or NVM Express (NVMe) is the next must-have storage technology. Let s look at what NVMe delivers. NVMe is a communications protocol
More informationIntroduction to I/O and Disk Management
Introduction to I/O and Disk Management 1 Secondary Storage Management Disks just like memory, only different Why have disks? Ø Memory is small. Disks are large. Short term storage for memory contents
More informationOperating Systems. Operating Systems Professor Sina Meraji U of T
Operating Systems Operating Systems Professor Sina Meraji U of T How are file systems implemented? File system implementation Files and directories live on secondary storage Anything outside of primary
More informationHP VMA-series Memory Arrays
HP VMA-series Memory Arrays Optimize OLTP database applications Presenter title August 2011 2011 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without
More informationCSE 451: Operating Systems Spring Module 12 Secondary Storage
CSE 451: Operating Systems Spring 2017 Module 12 Secondary Storage John Zahorjan 1 Secondary storage Secondary storage typically: is anything that is outside of primary memory does not permit direct execution
More informationCSE 451: Operating Systems Spring Module 12 Secondary Storage. Steve Gribble
CSE 451: Operating Systems Spring 2009 Module 12 Secondary Storage Steve Gribble Secondary storage Secondary storage typically: is anything that is outside of primary memory does not permit direct execution
More informationStorage: HDD, SSD and RAID
Storage: HDD, SSD and RAID Johan Montelius KTH 2017 1 / 33 Why? 2 / 33 Why? Give me two reasons why we would like to have secondary storage? 2 / 33 Computer architecture Gigabyte Z170 Gaming 2 PCIe x16/x4
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 informationCOS 318: Operating Systems. Storage Devices. Vivek Pai Computer Science Department Princeton University
COS 318: Operating Systems Storage Devices Vivek Pai Computer Science Department Princeton University http://www.cs.princeton.edu/courses/archive/fall11/cos318/ Today s Topics Magnetic disks Magnetic disk
More informationHow to Pick SQL Server Hardware
How to Pick SQL Server Hardware The big picture 1. What SQL Server edition do you need? 2. Does your RPO/RTO dictate shared storage? 3. If you need shared storage, what s important? 4. No-brainer answers
More informationMicron Quad-Level Cell Technology Brings Affordable Solid State Storage to More Applications
Micron Quad-Level Cell Technology Brings Affordable Solid State Storage to More Applications QLC Empowers Immense, Read-Focused Workloads Overview For years, read-focused workloads were relegated to legacy
More informationWhy? Storage: HDD, SSD and RAID. Computer architecture. Computer architecture. 10 µs - 10 ms. Johan Montelius
Why? Storage: HDD, SSD and RAID Johan Montelius Give me two reasons why we would like to have secondary storage? KTH 2017 1 / 33 Computer architecture 2 4 2 6 4 6 2 1 1 4 Computer architecture GPU Gigabyte
More informationOracle Exadata: Strategy and Roadmap
Oracle Exadata: Strategy and Roadmap - New Technologies, Cloud, and On-Premises Juan Loaiza Senior Vice President, Database Systems Technologies, Oracle Safe Harbor Statement The following is intended
More informationStorage: HDD, SSD and RAID
Storage: HDD, SSD and RAID Johan Montelius KTH 2017 1 / 33 Why? Give me two reasons why we would like to have secondary storage? 2 / 33 Computer architecture Gigabyte Z170 Gaming 2 4 2 6 4 6 2 1 1 4 PCIe
More informationDemystifying Storage Area Networks. Michael Wells Microsoft Application Solutions Specialist EMC Corporation
Demystifying Storage Area Networks Michael Wells Microsoft Application Solutions Specialist EMC Corporation About Me DBA for 7+ years Developer for 10+ years MCSE: Data Platform MCSE: SQL Server 2012 MCITP:
More informationu Covered: l Management of CPU & concurrency l Management of main memory & virtual memory u Currently --- Management of I/O devices
Where Are We? COS 318: Operating Systems Storage Devices Jaswinder Pal Singh Computer Science Department Princeton University (http://www.cs.princeton.edu/courses/cos318/) u Covered: l Management of CPU
More informationStorage Systems : Disks and SSDs. Manu Awasthi CASS 2018
Storage Systems : Disks and SSDs Manu Awasthi CASS 2018 Why study storage? Scalable High Performance Main Memory System Using Phase-Change Memory Technology, Qureshi et al, ISCA 2009 Trends Total amount
More informationStorage Update and Storage Best Practices for Microsoft Server Applications. Dennis Martin President, Demartek January 2009 Copyright 2009 Demartek
Storage Update and Storage Best Practices for Microsoft Server Applications Dennis Martin President, Demartek January 2009 Copyright 2009 Demartek Agenda Introduction Storage Technologies Storage Devices
More informationUCS Invicta: A New Generation of Storage Performance. Mazen Abou Najm DC Consulting Systems Engineer
UCS Invicta: A New Generation of Storage Performance Mazen Abou Najm DC Consulting Systems Engineer HDDs Aren t Designed For High Performance Disk 101 Can t spin faster (200 IOPS/Drive) Can t seek faster
More informationStorage Best Practices for Microsoft Server Applications
Storage Best Practices for Microsoft Server Applications Dennis Martin President, Demartek Principal Research Contributor Performance Lab Portal wikibon.org August 2009 1 Agenda Brief Company Overview
More informationIncreasing Performance of Existing Oracle RAC up to 10X
Increasing Performance of Existing Oracle RAC up to 10X Prasad Pammidimukkala www.gridironsystems.com 1 The Problem Data can be both Big and Fast Processing large datasets creates high bandwidth demand
More informationSecondary storage. CS 537 Lecture 11 Secondary Storage. Disk trends. Another trip down memory lane
Secondary storage CS 537 Lecture 11 Secondary Storage Michael Swift Secondary storage typically: is anything that is outside of primary memory does not permit direct execution of instructions or data retrieval
More informationStorage Technologies - 3
Storage Technologies - 3 COMP 25212 - Lecture 10 Antoniu Pop antoniu.pop@manchester.ac.uk 1 March 2019 Antoniu Pop Storage Technologies - 3 1 / 20 Learning Objectives - Storage 3 Understand characteristics
More informationSurFS Product Description
SurFS Product Description 1. ABSTRACT SurFS An innovative technology is evolving the distributed storage ecosystem. SurFS is designed for cloud storage with extreme performance at a price that is significantly
More informationHP visoko-performantna OLTP rješenja
HP visoko-performantna OLTP rješenja Tomislav Alpeza Presales Consultant, BCS/SD 2011 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice Performance
More information[537] Flash. Tyler Harter
[537] Flash Tyler Harter Flash vs. Disk Disk Overview I/O requires: seek, rotate, transfer Inherently: - not parallel (only one head) - slow (mechanical) - poor random I/O (locality around disk head) Random
More informationHYBRID STORAGE TM. WITH FASTier ACCELERATION TECHNOLOGY
HYBRID STORAGE TM WITH FASTier ACCELERATION TECHNOLOGY Nexsan s FASTier acceleration technology uses advanced software architecture and algorithms to leverage the power of solid-state to accelerate the
More informationAnnual Update on Flash Memory for Non-Technologists
Annual Update on Flash Memory for Non-Technologists Jay Kramer, Network Storage Advisors & George Crump, Storage Switzerland August 2017 1 Memory / Storage Hierarchy Flash Memory Summit 2017 2 NAND Flash
More informationSNIA Tutorial 1 A CASE FOR FLASH STORAGE HOW TO CHOOSE FLASH STORAGE FOR YOUR APPLICATIONS
SNIA Tutorial 1 A CASE FOR FLASH STORAGE HOW TO CHOOSE FLASH STORAGE FOR YOUR APPLICATIONS Dejan Kocic, NetApp Flash Memory Storage 2018 Welcome to SNIA Education Afternoon at Flash Memory Summit 2018
More informationFlash In the Data Center
Flash In the Data Center Enterprise-grade Morgan Littlewood: VP Marketing and BD Violin Memory, Inc. Email: littlewo@violin-memory.com Mobile: +1.650.714.7694 7/12/2009 1 Flash in the Data Center Nothing
More informationWhy Scale-Out Big Data Apps Need A New Scale- Out Storage
Why Scale-Out Big Data Apps Need A New Scale- Out Storage Modern storage for modern business Rob Whiteley, VP, Marketing, Hedvig April 9, 2015 Big data pressures on storage infrastructure The rise of elastic
More informationLecture 29. Friday, March 23 CS 470 Operating Systems - Lecture 29 1
Lecture 29 Reminder: Homework 7 is due on Monday at class time for Exam 2 review; no late work accepted. Reminder: Exam 2 is on Wednesday. Exam 2 review sheet is posted. Questions? Friday, March 23 CS
More informationSSD (Solid State Disk)
SSD (Solid State Disk) http://en.wikipedia.org/wiki/solid-state_drive SSD (Solid State Disk) drives Most SSD drives gives very good performance 4x ~ 100x No noise, low weight, power and heat generation
More informationCOS 318: Operating Systems. Storage Devices. Jaswinder Pal Singh Computer Science Department Princeton University
COS 318: Operating Systems Storage Devices Jaswinder Pal Singh Computer Science Department Princeton University http://www.cs.princeton.edu/courses/archive/fall13/cos318/ Today s Topics Magnetic disks
More informationUNIVERSITY OF MASSACHUSETTS Dept. of Electrical & Computer Engineering. Computer Architecture ECE 568
UNIVERSITY OF MASSACHUSETTS Dept. of Electrical & Computer Engineering Computer Architecture ECE 568 Part 6 Input/Output Israel Koren ECE568/Koren Part.6. Motivation: Why Care About I/O? CPU Performance:
More information- SLED: single large expensive disk - RAID: redundant array of (independent, inexpensive) disks
RAID and AutoRAID RAID background Problem: technology trends - computers getting larger, need more disk bandwidth - disk bandwidth not riding moore s law - faster CPU enables more computation to support
More informationLecture 16: Storage Devices
CS 422/522 Design & Implementation of Operating Systems Lecture 16: Storage Devices Zhong Shao Dept. of Computer Science Yale University Acknowledgement: some slides are taken from previous versions of
More informationFILE SYSTEMS, PART 2. CS124 Operating Systems Fall , Lecture 24
FILE SYSTEMS, PART 2 CS124 Operating Systems Fall 2017-2018, Lecture 24 2 Last Time: File Systems Introduced the concept of file systems Explored several ways of managing the contents of files Contiguous
More informationSFS: Random Write Considered Harmful in Solid State Drives
SFS: Random Write Considered Harmful in Solid State Drives Changwoo Min 1, 2, Kangnyeon Kim 1, Hyunjin Cho 2, Sang-Won Lee 1, Young Ik Eom 1 1 Sungkyunkwan University, Korea 2 Samsung Electronics, Korea
More informationIsilon Scale Out NAS. Morten Petersen, Senior Systems Engineer, Isilon Division
Isilon Scale Out NAS Morten Petersen, Senior Systems Engineer, Isilon Division 1 Agenda Architecture Overview Next Generation Hardware Performance Caching Performance SMB 3 - MultiChannel 2 OneFS Architecture
More informationSSDs Driving Greater Efficiency in Data Centers
SSDs Driving Greater Efficiency in Data Centers - Tutorial - June 23 rd, 2010 Tony Kim Samsung Semiconductor Inc. Table of Contents Part I: IT Industry Trend & SSD Market Outlook SSD Introduction & Application
More informationCOS 318: Operating Systems. Storage Devices. Kai Li Computer Science Department Princeton University
COS 318: Operating Systems Storage Devices Kai Li Computer Science Department Princeton University http://www.cs.princeton.edu/courses/archive/fall11/cos318/ Today s Topics Magnetic disks Magnetic disk
More informationCluster-Level Google How we use Colossus to improve storage efficiency
Cluster-Level Storage @ Google How we use Colossus to improve storage efficiency Denis Serenyi Senior Staff Software Engineer dserenyi@google.com November 13, 2017 Keynote at the 2nd Joint International
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 informationNew HPE 3PAR StoreServ 8000 and series Optimized for Flash
New HPE 3PAR StoreServ 8000 and 20000 series Optimized for Flash AGENDA HPE 3PAR StoreServ architecture fundamentals HPE 3PAR Flash optimizations HPE 3PAR portfolio overview HPE 3PAR Flash example from
More informationPRESENTATION TITLE GOES HERE
Enterprise Storage PRESENTATION TITLE GOES HERE Leah Schoeb, Member of SNIA Technical Council SNIA EmeraldTM Training SNIA Emerald Power Efficiency Measurement Specification, for use in EPA ENERGY STAR
More informationQuickSpecs. PCIe Solid State Drives for HP Workstations
Overview Introduction Storage technology with NAND media is outgrowing the bandwidth limitations of the SATA bus. New high performance Storage solutions will connect directly to the PCIe bus for revolutionary
More informationCISC 7310X. C11: Mass Storage. Hui Chen Department of Computer & Information Science CUNY Brooklyn College. 4/19/2018 CUNY Brooklyn College
CISC 7310X C11: Mass Storage Hui Chen Department of Computer & Information Science CUNY Brooklyn College 4/19/2018 CUNY Brooklyn College 1 Outline Review of memory hierarchy Mass storage devices Reliability
More informationSSD Applications in the Enterprise Area
SSD Applications in the Enterprise Area Tony Kim Samsung Semiconductor January 8, 2010 Outline Part I: SSD Market Outlook Application Trends Part II: Challenge of Enterprise MLC SSD Understanding SSD Lifetime
More informationdavidklee.net heraflux.com linkedin.com/in/davidaklee
@kleegeek davidklee.net heraflux.com linkedin.com/in/davidaklee Specialties / Focus Areas / Passions: Performance Tuning & Troubleshooting Virtualization Cloud Enablement Infrastructure Architecture Health
More informationUNIVERSITY OF MASSACHUSETTS Dept. of Electrical & Computer Engineering. Computer Architecture ECE 568
UNIVERSITY OF MASSACHUSETTS Dept. of Electrical & Computer Engineering Computer Architecture ECE 568 Part 6 Input/Output Israel Koren ECE568/Koren Part.6. CPU performance keeps increasing 26 72-core Xeon
More informationMass-Storage. ICS332 - Fall 2017 Operating Systems. Henri Casanova
Mass-Storage ICS332 - Fall 2017 Operating Systems Henri Casanova (henric@hawaii.edu) Magnetic Disks! Magnetic disks (a.k.a. hard drives ) are (still) the most common secondary storage devices today! They
More informationKey Points. Rotational delay vs seek delay Disks are slow. Techniques for making disks faster. Flash and SSDs
IO 1 Today IO 2 Key Points CPU interface and interaction with IO IO devices The basic structure of the IO system (north bridge, south bridge, etc.) The key advantages of high speed serial lines. The benefits
More informationAdvanced Database Systems
Lecture II Storage Layer Kyumars Sheykh Esmaili Course s Syllabus Core Topics Storage Layer Query Processing and Optimization Transaction Management and Recovery Advanced Topics Cloud Computing and Web
More informationSolid State Performance Comparisons: SSD Cache Performance
Solid State Performance Comparisons: SSD Cache Performance Dennis Martin, President, Demartek This presentation is available at http://www.demartek.com/demartek_presenting_snwusa_2013-10.html Agenda Demartek
More informationFile System Management
Lecture 8: Storage Management File System Management Contents Non volatile memory Tape, HDD, SSD Files & File System Interface Directories & their Organization File System Implementation Disk Space Allocation
More informationStorage Systems. Storage Systems
Storage Systems Storage Systems We already know about four levels of storage: Registers Cache Memory Disk But we've been a little vague on how these devices are interconnected In this unit, we study Input/output
More informationDisks. Storage Technology. Vera Goebel Thomas Plagemann. Department of Informatics University of Oslo
Disks Vera Goebel Thomas Plagemann 2014 Department of Informatics University of Oslo Storage Technology [Source: http://www-03.ibm.com/ibm/history/exhibits/storage/storage_photo.html] 1 Filesystems & Disks
More informationRef: Chap 12. Secondary Storage and I/O Systems. Applied Operating System Concepts 12.1
Ref: Chap 12 Secondary Storage and I/O Systems Applied Operating System Concepts 12.1 Part 1 - Secondary Storage Secondary storage typically: is anything that is outside of primary memory does not permit
More informationHow to Speed up Database Applications with a Purpose-Built SSD Storage Solution
How to Speed up Database Applications with a Purpose-Built SSD Storage Solution SAN Accessible Storage Array Speeds Applications by up to 25x Introduction Whether deployed in manufacturing, finance, web
More informationSQL Server 2014 Upgrade
SQL Server 2014 Upgrade Case study featuring In-Memory OLTP and Hybrid-Cloud Scenarios Evgeny Ternovsky, Program Manager II, Data Platform Group Bill Kan, Service Engineer II, Data Platform Group Background
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 informationIBM Emulex 16Gb Fibre Channel HBA Evaluation
IBM Emulex 16Gb Fibre Channel HBA Evaluation Evaluation report prepared under contract with Emulex Executive Summary The computing industry is experiencing an increasing demand for storage performance
More informationBuilding an All Flash Server What s the big deal? Isn t it all just plug and play?
Building an All Flash Server What s the big deal? Isn t it all just plug and play? Doug Rollins Micron Technology Santa Clara, CA 1 What we ll cover Industry Secrets (shhhhh. ) Example Platform Key features
More informationOracle Exadata. Smart Database Platforms - Dramatic Performance and Cost Advantages. Juan Loaiza Senior Vice President Oracle Database Systems
Oracle Exadata Smart Database Platforms - Dramatic Performance and Cost Advantages Juan Loaiza Senior Vice President Oracle Database Systems Exadata X5-2 Exadata X5-8 SuperCluster M7-8 Exadata Vision Dramatically
More informationAchieving Memory Level Performance: Secrets Beyond Shared Flash
Achieving Memory Level Performance: Secrets Beyond Shared Flash Kothanda (Kodi) Umamageswaran Vice President, Exadata Development Gurmeet Goindi Exadata Product Management Safe Harbor Statement The following
More informationOptimizing Server Designs for Speed
Optimizing Server Designs for Speed Optimizing Server Designs for Speed We will discuss the latest in server hardware, virtualization, and disk storage that boosts Skyward s performance. We will also discuss
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 informationIBM DS8880F All-flash Data Systems
IBM DS8880F All-flash Data Systems Gary F Albert Offering Manager and Business Line Manager IBM DS8880 Release 8..1 and Roadmap November 016 NAND Flash technology is used across the IBM Systems Flash Storage
More informationWas ist dran an einer spezialisierten Data Warehousing platform?
Was ist dran an einer spezialisierten Data Warehousing platform? Hermann Bär Oracle USA Redwood Shores, CA Schlüsselworte Data warehousing, Exadata, specialized hardware proprietary hardware Introduction
More informationDescribe the features and performance characteristics of server-class hard disks and solid state drives.
Storage Devices Storage Devices Objectives On completion of this unit, you will be able to: Describe the features and performance characteristics of server-class hard disks and solid state drives. Describe
More informationComputer Architecture 计算机体系结构. Lecture 6. Data Storage and I/O 第六讲 数据存储和输入输出. Chao Li, PhD. 李超博士
Computer Architecture 计算机体系结构 Lecture 6. Data Storage and I/O 第六讲 数据存储和输入输出 Chao Li, PhD. 李超博士 SJTU-SE346, Spring 2018 Review Memory hierarchy Cache and virtual memory Locality principle Miss cache, victim
More informationIBM FlashSystem. IBM FLiP Tool Wie viel schneller kann Ihr IBM i Power Server mit IBM FlashSystem 900 / V9000 Storage sein?
FlashSystem Family 2015 IBM FlashSystem IBM FLiP Tool Wie viel schneller kann Ihr IBM i Power Server mit IBM FlashSystem 900 / V9000 Storage sein? PiRT - Power i Round Table 17 Sep. 2015 Daniel Gysin IBM
More informationFusion iomemory PCIe Solutions from SanDisk and Sqrll make Accumulo Hypersonic
WHITE PAPER Fusion iomemory PCIe Solutions from SanDisk and Sqrll make Accumulo Hypersonic Western Digital Technologies, Inc. 951 SanDisk Drive, Milpitas, CA 95035 www.sandisk.com Table of Contents Executive
More informationPractical MySQL Performance Optimization. Peter Zaitsev, CEO, Percona July 20 th, 2016 Percona Technical Webinars
Practical MySQL Performance Optimization Peter Zaitsev, CEO, Percona July 20 th, 2016 Percona Technical Webinars In This Presentation We ll Look at how to approach Performance Optimization Discuss Practical
More informationHigh Performance SSD & Benefit for Server Application
High Performance SSD & Benefit for Server Application AUG 12 th, 2008 Tony Park Marketing INDILINX Co., Ltd. 2008-08-20 1 HDD SATA 3Gbps Memory PCI-e 10G Eth 120MB/s 300MB/s 8GB/s 2GB/s 1GB/s SSD SATA
More informationMass-Storage. ICS332 Operating Systems
Mass-Storage ICS332 Operating Systems Magnetic Disks Magnetic disks are (still) the most common secondary storage devices today They are messy Errors, bad blocks, missed seeks, moving parts And yet, the
More informationInfrastructure Tuning
Infrastructure Tuning For SQL Server Performance SQL PASS Performance Virtual Chapter 2014.07.24 About David Klee @kleegeek davidklee.net gplus.to/kleegeek linked.com/a/davidaklee Specialties / Focus Areas
More informationFlash memory talk Felton Linux Group 27 August 2016 Jim Warner
Flash memory talk Felton Linux Group 27 August 2016 Jim Warner Flash Memory Summit Annual trade show at Santa Clara Convention Center Where there is money, trade shows follow. August 8 11, 2016 Borrowing
More informationDDN. DDN Updates. DataDirect Neworks Japan, Inc Nobu Hashizume. DDN Storage 2018 DDN Storage 1
1 DDN DDN Updates DataDirect Neworks Japan, Inc Nobu Hashizume DDN Storage 2018 DDN Storage 1 2 DDN A Broad Range of Technologies to Best Address Your Needs Your Use Cases Research Big Data Enterprise
More informationChapter 10: Mass-Storage Systems. Operating System Concepts 9 th Edition
Chapter 10: Mass-Storage Systems Silberschatz, Galvin and Gagne 2013 Chapter 10: Mass-Storage Systems Overview of Mass Storage Structure Disk Structure Disk Attachment Disk Scheduling Disk Management Swap-Space
More informationSSDs vs HDDs for DBMS by Glen Berseth York University, Toronto
SSDs vs HDDs for DBMS by Glen Berseth York University, Toronto So slow So cheap So heavy So fast So expensive So efficient NAND based flash memory Retains memory without power It works by trapping a small
More informationA Prototype Storage Subsystem based on PCM
PSS A Prototype Storage Subsystem based on IBM Research Zurich Ioannis Koltsidas, Roman Pletka, Peter Mueller, Thomas Weigold, Evangelos Eleftheriou University of Patras Maria Varsamou, Athina Ntalla,
More informationWhite Paper Features and Benefits of Fujitsu All-Flash Arrays for Virtualization and Consolidation ETERNUS AF S2 series
White Paper Features and Benefits of Fujitsu All-Flash Arrays for Virtualization and Consolidation Fujitsu All-Flash Arrays are extremely effective tools when virtualization is used for server consolidation.
More informationBenefits of Automatic Data Tiering in OLTP Database Environments with Dell EqualLogic Hybrid Arrays
TECHNICAL REPORT: Performance Study Benefits of Automatic Data Tiering in OLTP Database Environments with Dell EqualLogic Hybrid Arrays ABSTRACT The Dell EqualLogic hybrid arrays PS6010XVS and PS6000XVS
More informationDDN. DDN Updates. Data DirectNeworks Japan, Inc Shuichi Ihara. DDN Storage 2017 DDN Storage
DDN DDN Updates Data DirectNeworks Japan, Inc Shuichi Ihara DDN A Broad Range of Technologies to Best Address Your Needs Protection Security Data Distribution and Lifecycle Management Open Monitoring Your
More informationA Deeper Dive: Where (And How and Why) to Implement Solid-State Storage
A Deeper Dive: Where (And How and Why) to Implement Solid-State Storage Dennis Martin President, Demartek Storage Decisions 2012 TechTarget Solid-State Storage Agenda About Demartek Solid-state storage
More informationSSD Server Hard Drives for Dell
New for 2011! Axiom is introducing a new line of SSD Hot-Swap Server Hard Drives. Our high performance SSD hard drives have been paired with system specific Hot-Swap trays to deliver seamless integration
More information