Intelligent Storage Systems 1 What is an Intelligent

Intelligent Storage Systems 1

What is an Intelligent Storage Systems are RAID arrays that are: • Highly optimized for I/O processing • Have large amounts of cache for improving I/O performance • Have operating environments that provide: – Intelligence for managing cache – Array resource allocation – Connectivity for heterogeneous hosts – Advanced array based local and remote replication options ISMDR: BE 5 IT: VIII: Madhu N. PIIT 2

Benefits of an Intelligent Storage System Intelligent storage system provides several benefits over a collection of disks in an array (JBOD) or even a RAID arrays: – Increased capacity – Improved performance – Easier data management – Improved data availability and protection – Enhanced Business Continuity support – Improved security and access control ISMDR: BE 5 IT: VIII: Madhu N. PIIT 3

Components of an Intelligent Storage System Host Front End Back End Physical Disks Cache Connectivity FC SAN ISMDR: BE 5 IT: VIII: Madhu N. PIIT 4

Intelligent Storage System: Front End Intelligent Storage System Host Front End Back End Physical Disks Cache Connectivity FC SAN Ports Controllers ISMDR: BE 5 IT: VIII: Madhu N. PIIT 5

Front End • provides the interface between the storage system and the host • two components: front-end ports and front-end controllers • ports enable hosts to connect to the intelligent storage system- processing logic that executes the appropriate transport protocol- high availability. • controllers route data to and from cache receives write data, the controller sends an acknowledgment message. optimize I/O processing by using command queuing algorithms.

Front-End Command Queuing • determines the execution order of received commands and can reduce unnecessary drive head movements and improve disk performance • Command -- controller(alg)-- assigns tag(seq of execution) • First In First Out (FIFO) • Seek Time Optimization • Access Time optimization

Front End Command Queuing A D C B A D I/O Requests C B A I/O Processing Order C D B Front-End Controller Cylinders Without Optimization (FIFO) A D C B A D I/O Requests B C A I/O Processing Order Front-End Controller With command queuing ISMDR: BE 5 IT: VIII: Madhu N. PIIT C D B Cylinders 8

Intelligent Storage System: Cache Intelligent Storage System Front End Host Back End Physical Disks Cache Connectivity FC SAN ISMDR: BE 5 IT: VIII: Madhu N. PIIT 9

Read Operation with Cache: ‘Hits’ and ‘Misses’ Data found in cache = ‘Hit’ Read Request Cache No data found = ‘Miss’ Cache Read Request ISMDR: BE 5 IT: VIII: Madhu N. PIIT 10

Write Operation with Cache Write-through Cache Write Request Acknowledgement Write-back Cache Write Request Acknowledgement ISMDR: BE 5 IT: VIII: Madhu N. PIIT 11

Write Operation with Cache • Write aside Size-bypassing cache. • provide performance advantages ISMDR: BE 5 IT: VIII: Madhu N. PIIT 12

Cache Management: Algorithms New Data • finite and expensive resource that needs proper management • Least Recently Used (LRU) – Discards least recently used data • Most Recently Used (MRU) – Discards most recently used data ( assumed that recent data may not be required for a while ) Oldest Data Cache Implementation Dedicated Cache – Separate memory sets reserved for read and write Global Cache – Both read and write operation use available memory. – More efficient- only one global set of addresses has to be managed. ratio of cache available for reads versus writes is dynamically adjusted based on the workloads. ISMDR: BE 5 IT: VIII: Madhu N. PIIT 13

Cache Management: Watermarking I/O access rate and pattern, high and low levels called watermarks are set in cache to manage the flushing Manage peak I/O requests “bursts” through flushing/de-staging. Flushing is the process of committing data from cache to the disk. – Idle flushing, High Watermark flushing and Forced flushing For maximum performance: – Provide headroom in write cache for I/O bursts 100 % HWM LWM Idle flushing High watermark flushing Forced flushing ISMDR: BE 5 IT: VIII: Madhu N. PIIT

Cache Data Protection • Protecting cache data against failure: (power failure) – Cache mirroring • Each write to the cache is held in two different memory locations on two independent memory cards • Cache coherency(only writes r mirrored) – Cache vaulting • Cache is exposed to the risk of uncommitted data loss due to power failure • In the event of power failure, uncommitted data is dumped to a dedicated set of drives called vault drives ISMDR: BE 5 IT: VIII: Madhu N. PIIT 15

Intelligent Storage System: Back End Intelligent Storage System Host Front End Back End Physical Disks Cache Connectivity FC SAN Controllers ISMDR: BE 5 IT: VIII: Madhu N. PIIT Ports 16

Intelligent Storage System: Physical Disks Intelligent Storage System Host Front End Back End Physical Disks Cache Connectivity FC SAN ISMDR: BE 5 IT: VIII: Madhu N. PIIT 17

What the Host Sees – RAID Sets and LUNs (Logical Unit Number) Host 1 Intelligent Storage System Front End LUN 0 Connectivity Back End Cache Physical Disks LUN 0 FC SAN LUN 1 Host 2 ISMDR: BE 5 IT: VIII: Madhu N. PIIT 18

LUN Masking It is a process that provides data access control by defining which LUNs a host can access. • LUN masking is access control mechanism • Process of masking LUNs from unauthorized access • Implemented on storage arrays and at front end controller Intelligent Storage Arrays High End Storage System Mid Range Storage System ISMDR: BE 5 IT: VIII: Madhu N. PIIT 19

• ISS Implementation: High-end Storage Systems Also referred as Active-active arrays – I/O’s are serviced through all the available path Active-Active Configuration Active Host Port Controller B Large storage capacity Huge cache to service host I/Os Fault tolerance architecture Multiple front-end ports and support to interface protocols – High scalability – Ability to handle large amounts of concurrent I/Os – – Controller A • Following are high-end array capabilities: LUN Storage Array • Designed for large enterprises ISMDR: BE 5 IT: VIII: Madhu N. PIIT 20

Midrange Storage Systems Active Port Passive Host ISMDR: BE 5 IT: VIII: Madhu N. PIIT Controller A – Host can perform I/Os to LUNs only through active paths – Other paths remain passive till active path fails • Midrange array have two controllers, each with cache, RAID controllers and disks drive interfaces • Designed for small and medium enterprises • Less scalable as compared to high-end array Active-Passive Configuration Controller B • Also referred as Active-passive arrays LUN Storage Array

Chapter Summary Key points covered in this chapter: • Intelligent Storage Systems features • Components of Intelligent Storage Systems • Cache management algorithms • Intelligent Storage System implementation – High-end storage array – Mid range storage array ISMDR: BE 5 IT: VIII: Madhu N. PIIT 22

Check Your Knowledge • What are the parts of an Intelligent Storage System? • What are the differences between a high-end and midrange storage array? • What is the difference between a read cache hit and a read cache miss? • What is the difference between Least Recently Used and Most Recently Used algorithms? • What is the difference between Write-through and Write-back cache? ISMDR: BE 5 IT: VIII: Madhu N. PIIT 23