Manage rising disk prices with storage virtualization webinar
Download as PPTX, PDF0 likes1,823 views
This document discusses strategies for managing rising disk prices through storage virtualization, emphasizing the importance of reclaiming existing storage, improving utilization, and deferring capital purchases. It outlines various technologies such as thin provisioning, dynamic tiering, and storage reclamation techniques to enhance storage efficiency. Additionally, it promotes upcoming educational sessions to help customers leverage these technologies for cost savings and better business performance.
Manage rising disk prices with storage virtualization webinar
- 1. MANAGE RISING DISK PRICES WITH STORAGE VIRTUALIZATION JANUARY 25, 2012 Steve Burr, Solution Architect, Services Engineering Mike Nalls, Senior Product Marketing Manager
- 2. WEBTECH EDUCATIONAL SERIES Manage Rising Disk Prices with Storage Virtualization Technical Session: Learn how storage virtualization can reclaim existing storage on the floor. Extend thin provisioning to existing storage to increase disk utilization and defer capital purchases. Take advantage of zero page reclaim and write same to reclaim storage reclamation. Use the Hitachi Dynamic Provisioning overprovisioning capability for automatic optimum storage utilization. We’ll also cover dramatic enhancements to the Hitachi Switch IT On III program that make this extremely attractive and affordable. Come to this WebTech and learn from Hitachi experts how to use these technologies to increase your customer satisfaction and sales despite impending increases in disk prices. Attend this session to learn Thin provisioning options for storage reclamation and expectations that depend on customer environments How to reclaim storage from existing systems using storage virtualization Details of the Hitachi Switch IT On III enhanced program
- 3. UPCOMING WEBTECH SESSIONS February The Economics of Storage Virtualization, February 1, 9 a.m. PT, 12 p.m. ET Reduce Costs by Eliminating Backups and Dependence on Local Storage at the Edge, February 8, 9 a.m. PT, 12 p.m. ET Improve Cloud Adoption with Converged Infrastructure Solutions from Hitachi Data Systems and Brocade, February 22, 9 a.m. PT, 12 p.m. ET March Lower Your Virtual Machine Costs with an Economical New Framework, March 14, 9 a.m. PT, 12 p.m. ET Flexible Options to Simplify Cloud Deployments, March 28, 9 a.m. PT, 12 p.m. ET Please check www.hds.com/webtech for Links to the recording, the presentation and Q&A (available next week) Schedule and registration for upcoming WebTech sessions
- 4. MANAGE RISING DISK PRICES WITH STORAGE VIRTUALIZATION JANUARY 25, 2012 Steve Burr, Solution Architect, Services Engineering Mike Nalls, Senior Product Marketing Manager
- 5. AGENDA • Capacity efficiency problem • Solutions • How-tos • Successes • Services and promotions • Q&A session IT IS CHEAPER FOR YOU TO VIRTUALIZE AND RECLAIM THAN TO PURCHASE THAT CAPACITY
- 6. CAPACITY EFFICIENCY PROBLEM: DISK PRICES AND SHORTAGES Unprecedented geological event causing worldwide disk shortage, resulting in Limited availability and increased price of HDD Compromised growth and business disruption Forecasting challenges Missed sales and revenue targets Shortage may persist into 2013, creating immediate need for increased capacity efficiency worldwide, via Increased storage utilization Capacity purchase deferral Capacity reclamation Extension of useful life of existing assets
- 7. THE FOUNDATION FOR CAPACITY EFFICIENCY Increase storage utilization - Use more of existing and new assets Capacity purchase deferral - Reduce need to purchase capacity in step with business growth Capacity reclamation - Free up high-performance capacity for future use Extend useful life of existing assets - Reduce migration period to allow for extended use of systems Repurpose existing assets - Manage heterogeneous storage capacity as a single pool that inherits the benefits of parent array Take advantage of cost savings features, such as thin provisioning and dynamic data mobility, with existing storage systems, not just new ones
- 8. CAPACITY-EFFICIENCY SOLUTIONS REDUCE NEED, RECLAIM EXISTING CAPACITY AND DEFER PURCHASES ‒ Storage pooling and thin provisioning ‒ Virtualization ‒ Tiered storage and dynamic tiering ‒ Disk configuration optimization: RAID 5 and 6 vs. RAID 10; high vs. low density ‒ Storage reclamation: ZPR and write same ‒ Thin replication ‒ Copy-on-write ‒ Service catalogs and chargeback Information ‒ Intelligent archiving (compression, deduplication and single instancing)
- 9. STORAGE POOLING AND DYNAMIC PROVISIONING Thin provisioning reduces wasted allocated capacity, provides improved performance with wide stripping, and enables storage reclamation Host Servers Dynamic Provisioning Thin Provisioned Volume Volumes (Virtual LUN) Actual storage Dynamic Provisioning capacity is Pool assigned only when host writes data to a LUN LDEVs LDEV LDEV LDEV LDEV LDEV LDEV LDEV LDEV Array Groups/Disk Drives
- 10. STORAGE VIRTUALIZATION Virtualization infrastructure impact NAS Mainframe VTL Virtual CAS Aggregates enterprise storage services for Servers agile, scalable, service level-based delivery Reduces complexity, reclaims capacity, simplifies management, improves integrity and reliability, lowers operations costs Leverages existing investments, improves ROA Information storage impact Virtualized Storage Solves content problems, provides application and media independence Virtualization outside the box Enables seamless integration of new and future capabilities Enables cloud storage, data migrations, lifecycle management, media migration, integration of data across applications
- 11. TIERED STORAGE, DYNAMIC TIERING AND AUTO TIERING Pool Data Heat Index Tier 0 High Activity Set Least Tier 1 Referenced Pages Normal Working Set Tier 2 Dynamic Tiering Volume Quiet • Dynamic provisioning saves capacity Data Set • Dynamic tiering optimizes cost and utilization • Choose tier policy on each LUN • Finer control is available through HDT auto-tiering solutions
- 12. DISK CONFIGURATION OPTIMIZATION With traditional volumes, Raid 5 and 6 have the power for all but the most demanding applications With Hitachi Dynamic Provisioning, wide striping often allows you to use raid levels with higher efficiency in more scenarios With Hitachi Dynamic Provisioning, wide striping may allow you to consider higher density drives With Hitachi Dynamic Tiering, optimum storage automatically delivers the most cost-effective use of your assets
- 13. ZERO PAGE RECLAIM Returns unused storage to pool Runs on storage, scans virtual volume for zero pages, returns any empty pages to pool free space Cannot know where “data” is a deleted file Physical Volume Virtual Volume V-VOL Server volume
- 14. WRITE SAME Returns unused storage to pool Runs on server, scans file system for free space and deleted files, sends command to return pages to pool free space Only server software can know if “data” is a deleted file Physical Volume Virtual Volume V-VOL Server volume Two scenarios • Veritas Storage Foundation • VMWare VAAI
- 15. THIN REPLICATION All Hitachi replication is thin-aware Savings made at the primary are also made at the secondary Less bandwidth Less capacity Distance replication capacity reclaimed
- 16. COPY-ON-WRITE Minimal use of disk space - Copies only changed data to a copy-on-write pool - Shares the date between multiple generations of snapshots Unchanged data never copied PVOL COW POOL ... VVOL Snapshot 1 time PVOL ... ... VVOL Snapshot 2 PVOL Data changed once shared 16
- 17. SERVICE CATALOGS AND CHARGEBACK INFORMATION Control user behavior with service catalogs, chargeback and showback reports, and basic consumption metrics
- 18. SERVICE CATALOGS AND CHARGEBACK INFORMATION Control user behavior with service catalogs, chargeback and showback reports, and basic consumption metrics
- 19. INTELLIGENT ARCHIVE SOLUTIONS Can cut down backup costs and move stale data to a lower- cost tier for retention or immutability requirements - Compression (Hitachi Content Platform) - Deduplication and single instancing (HCP − especially useful on archival storage because data is less active and return on deduplication is high)
- 21. SUCCESSFUL STORAGE RECLAIM We’ve been making recommendations for thin- provisioning best practices since first launching Hitachi Dynamic Provisioning But “best practice” implies you can ‒ start from a blank sheet ‒ strip everything down and start again Real-life environments migrate and evolve, best practices change and mistakes are made Global Solution Services storage reclaim service ‒ employs techniques developed to recover storage ‒ actively seeks opportunities to recover storage
- 22. OPPORTUNITIES TO RECOVER STORAGE But only where it make sense You can achieve zero waste with unlimited effort • One extreme – strip it all out and build it again • The other extreme – “opportunistic” zero page reclaim Consider the cost and benefit • One day to recover 50% of 100GB – probably not worth it • One day to recover 5% of 10TB – worth evaluating Cost includes management and outage
- 23. TECHNIQUES TO RECLAIM STORAGE Different for every file system, operating system and application − combinations must be considered Many of you have discovered some techniques Containers: Data is stored in nested containers like Russian dolls Example: Oracle ASM − a table, in a tablespace, in a data file, in a DISKGROUP, in DISK(s), in LUN(s)
- 24. SOME RECLAIM SCENARIOS Data LUN − don’t waste effort Data << LUN − may work… Data << LUN − may not work − deleted data • Can you zero data; will zero work? • Some approaches recover a lot but take time Do you understand all the containers?
- 25. SOME RECLAIM SCENARIOS Older file system that hogs all the space even when empty (e.g., UFS, HFS) Some reclamation but rebuild may be the only option But is this cost effective? Nothing here has impact Sometimes the problem is deep For maximum impact, engage all levels and classes of administration This can be a challenge Tablespace or table
- 26. EXAMPLE: ORACLE ASM Rows Deleted Tablespace Shrunk ASM Reclaimed Try to shrink ASM and you hit a challenge with ASM layer PDFs on www.hds.com ‒ Using ASM Storage Reclamation Utility from Oracle and Zero Page Reclamation Utility from Hitachi to Reclaim Storage Capacity ‒ Reclaiming Unused Storage Capacity on a Hitachi Virtual Storage Platform with ASM Storage Reclamation Utility from Oracle and Zero Page Reclamation Utility from Hitachi
- 29. STORAGE VIRTUALIZATION LEADER External >19,000 virtualization controllers shipped Storage >40% of customers actively use external storage Virtualization virtualization Leader Best infrastructure migration solution in the world Storage 2010 Hitachi Dynamic Provisioning penetration rate >50% Virtualization >500PB storage under management, 50% virtualized Management One enterprise platform for all data Leader
- 30. THE PROOF IS WITH OUR CUSTOMERS “With Hitachi virtualization technologies, we've seen storage capacity savings of 50% on some arrays, now provision storage in 25% of the time, and have increased utilization rates by over 30%. Overall, we've reduced our capital and operating costs for an improved return on our storage investment”. − Carter Lee Vice President of IT Operations, Overstock.com
- 31. STORAGE VIRTUALIZATION ECONOMICS: OVERSTOCK.COM Element Before After Number of Storage Arrays 8 4 Usable Capacity in TB 133 236 Tier 1 (75%) RAID 10 Tier 1 5% RAID 10 Tiers, RAID Tier 2 RAID 5 Tier 2 55% RAID 5 Software tiering Tier 3 40% RAID 6 Management 3 people, planning to hire 4th 3 people Migration Very disruptive, time consuming and complicated Done in 30 minutes Provisioning time 2-3 hours per incident 30 minutes per incident Volumes Thick Thin, ZPR in the future Overstock Unit Cost per Year $35 400,000 $30 Annual Capacity in GB 300,000 $25 TCO / GB / Year $20 200,000 $15 $10 100,000 $5 $0 0 2008 2009 2010 (est.) Project Cost IT Infrastructure Cost Business Burden
- 32. THE PROOF IS WITH OUR CUSTOMERS HUK Coburg reclaimed 30% capacity on IBM DS4800 using dynamic provisioning A global financial organization reclaimed 40-65% of its capacity through dynamic provisioning, saving in excess of US$4M Fidelity National reclaimed their entire investment in less than a year with a ROI of 188% (even after paying for the software)
- 33. CASE STUDY: LARGE CONSUMER PRODUCTS FIRM
- 34. CASE STUDY: MEDIUM-SIZED HEALTH CARE FIRM
- 35. CASE STUDY: COMPUTER SERVICES COMPANY
- 36. CASE STUDY: GLOBAL 500 INSURANCE COMPANY
- 37. WAYS TO HELP DRIVE EFFICIENCY Storage reclamation service ‒ Assessment tools and services to gauge reclamation potential Customer offer: Switch IT On III ‒ Available to both current and new VSP customers for virtualizing 3rd-party storage ‒ Includes incentives for value-added software
- 38. SWITCH IT ON III ENHANCED OFFER FOR VIRTUAL STORAGE PLATFORM Through June 30, 2012, when you purchase a new Hitachi Virtual Storage Platform, or for an existing system that is not already virtualizing external storage, in order to virtualize external storage behind it you can Get a 100TB capacity Basic Operating System (BOS) license block for the price of 50TB1 Offer applies to both software license and maintenance With the program, required BOS V license is discounted 75% Minimum of 25TB of external storage must be attached And you can “super-size” the offer with matching 100TB capacity license blocks at the price of 50TB of capacity1 for any of the following Hitachi Dynamic Tiering Hitachi Tuning Manager Hitachi Tiered Storage Manager Hitachi Command Director Also, under the program if you add the following products on the VSP, you can get them at a 50% reduction in price Hitachi Disaster Recovery bundle Hitachi In-System Replication bundle 1 conditions apply – see www.hds.com/go/free-storage-virtualization for additional details
- 39. IT IS CHEAPER FOR YOU TO VIRTUALIZE AND RECLAIM THAN TO PURCHASE THAT CAPACITY Investing in storage virtualization through VSP adds another dimension to storage efficiency by extending these new capabilities to existing storage systems If your current storage system does not provide thin provisioning, which can reclaim 20-40% or more of allocated but unused capacity, you don’t need to rip and replace it Just by attaching it behind VSP, VSP can see your existing LUNs and move them into a dynamic provisioning pool where the unused pages in the LUN can be reclaimed while your application is running
- 41. UPCOMING WEBTECH SESSIONS February The Economics of Storage Virtualization, February 1, 9 a.m. PT, 12 p.m. ET Reduce Costs by Eliminating Backups and Dependence on Local Storage at the Edge, February 8, 9 a.m. PT, 12 p.m. ET Improve Cloud Adoption with Converged Infrastructure Solutions from Hitachi Data Systems and Brocade, February 22, 9 a.m. PT, 12 p.m. ET March Lower Your Virtual Machine Costs with an Economical New Framework, March 14, 9 a.m. PT, 12 p.m. ET Flexible Options to Simplify Cloud Deployments, March 28, 9 a.m. PT, 12 p.m. ET Please check www.hds.com/webtech for: Links to the recording, the presentation and Q&A (available next week) Schedule and registration for upcoming WebTech sessions
- 42. THANK YOU
Editor's Notes
- #12: With Hitachi Dynamic Tiering software the complexities and overhead of implementing data lifecycle management and optimizing use of tiered storage are solved. Dynamic Tiering software simplifies storage administration by eliminating the need for time consuming manual data classification and movement of data to optimize usage of tiered storage.Hitachi Dynamic Tiering automatically moves data on fine-grain pages within Dynamic Tiering virtual volumes to the most appropriate media according to workload to maximize service levels and minimize TCO of storage. For example, a database index that is frequently read and written will migrate to high performance flash technology while older data that hasn’t been touched for a while will move to slower, cheaper disks. No elaborate decision criteria are needed; data is automatically moved according to simple rules. One, two or three tiers of storage can be defined and used within a single virtual volume using any of the storage media types available for the Hitachi Virtual Storage Platform. Tier creation is automatic based on user configuration policies, including media type and speed, RAID level and sustained I/O level requirements. Using ongoing embedded performance monitoring and periodic analysis the data is moved at a fine grain sub-LUN level to the most appropriate tier. The most active data moves to the highest tier. During the process the system automatically maximizes the use of storage keeping the higher tiers fully utilized.
- #19: What is being shown is the ability to produce chargeback data reports for block and file. HCS Command Director produces the information the customer needs to input into their chargeback systems. Usually customers have developed or selected some chargeback system covering their IT infrastructure, and want us to provide this needed information in tabular formats.
- #30: External Storage Virtualization Leader. These numbers are from a set of 8000 systems we were able to track. Using these we established parcentagees which were extrapolated as percentages of the larger total set of systems sold.Over 40% of our Enterprise Storage Systems customers have bought into our external storage virtualization strategy. Customers keep “switching it on”! It’s heterogeneous external storage virtualization too: a third of our external virtualized storage is third party and two thirds HDS AMS. Other vendors like IBM with SVC may claim many nodes but they don’t scale in capacity at all, and their virtualization technology cannot be undone once in place. That’s vendor lock-in and aquestionable architecture in a world where we will be measure in Zettabytes soon, if not already . Best infrastructure etc..External storage virtualization is one of the mechanisms we use in combination with other technologies to provide the “Best infrastructure migration solution”. Competitive :Our GSS teams are very experienced and successful. The technology and our skills are so good that we have many migrations under our belt of EMC to HDS migrations. Take them out of their DMX misery…or is it V-max? Also, our technology mitigates migration risk by limiting the outage time, or coming up soon, making it a seamless migration process. Storage Virtualization Management LeaderNeed to define “storage virtualization”: external + HDP and now HDT. I.e. not just external. HDP rate: (not for commentary, but more FYI: this 2010 snapshot represent the average of USP V/VM penetration (>30%) and VSP penetration (100%) )FYI500PB represent raw capacity under management for the USP V/VM; If questioned: 50% virtualization is the combination of HDP and external storage. We have 110-120 PB externally virtualized. Our numbers do not count short term connections for migrations. One enterprise platform for all data - Identify the fact that our 3D architecture, solutions with HNAS and HCAP provide a unique tool set for enterprise customers. No one else can do this, no one else WILL do this in the enterprise because of our IP investments and leadership etc.