virtual storage platform virtualization, dynamic ... gse session 20120314.pdf · 3d scaling...

Virtual Storage Platform (VSP) Update Virtualization, Dynamic Provisioning, and Dynamic Tiering for the Mainframe Stefaan Lemaire 14/03/2012

Technology Strategy

A common virtualized platform for all data, content and information

Structured Unstructured Semi-structured Rich Media

UNIFIED MANAGEMENT

VIRTUALIZED INFRASTRUCTURE

Intelligent

Virtual

Storage

Systems

Scalability

Advanced

Midrange

Storage

Systems

File and

Content

Storage

Functionality

HUS130 HUS110 HUS 150

VSP

Hitachi Virtual Storage Platform

Hitachi Unified Storage

Hitachi NAS Hitachi Content

Platform

Hitachi Data

Discovery Suite

Integrated Scaleable Storage Portfolio

AGENDA

Virtual Storage Platform – Quick Technology Update

Storage Virtualization in the Mainframe Environment

Dynamic Provisioning – Mainframe Environment

Dynamic Tiering – New for the Mainframe Environment

Virtual Storage Platform (VSP) – Other NEW Features

Summary

3D Scaling Architecture

5th generation network

crossbar switch

New control memory

architecture

2.4x OLTP performance

Greater connectivity

and choice

79% increase in

bandwidth

Scale up, scale out,

scale deep growth path

2x single image global

cache

40+% less power

requirements

COMPARISON: UNIVERSAL STORAGE PLATFORM V FIBRE CHANNEL LOOPS VS. VIRTUAL STORAGE PLATFORM SAS

Controller 1

4 – SAS

Links

Controller 0

4 – SAS

Links

4 – SAS

Links

4 – SAS

Links

Expander

Expander

En

clo

su

res

15

Dis

ks

SAS

HDD

Slot 0

SAS

HDD

Slot 1

SAS

HDD

Slot 2

SAS

HDD

Slot 3

SAS

HDD

Slot 4

SAS

HDD

Slot 13

SAS

HDD

Slot 14

Universal Storage Platform V with 64 Fibre

Channel (FC) loops 4 FC loops per

controller

4 FC loops per back end director

Half duplex arbitrated loops

32 FC loops pair at 4 Gb/s

Virtual Storage Platform with 64 SAS links

8 SAS paths per back end director blade

− 32 full duplex links per control chassis

− 64 SAS links at 6 Gb/s

Chassis are connected point-to-point

− Flash, SAS and SATA can be intermixed

Virtual Storage Platform

Capacity 2.5 PB of Internal Storage

255 PB of virtualized heterogeneous storage

1280 3.5” drives or 2048 2.5” drives or a combination of both sizes

65,536 logical devices (LDEV)

Drive Types High Performance SSD flash drives

SAS drives (2.5”)

SAS drives (3.5”) - Nearline

SATA drives (3.5”)

Connectivity (ports)

192 Fibre Channel

192 FICON

96 FCoE

New Drive Types and Capacities

Current Drive Technology:

‒ 2.5” 10K 300GB SAS

‒ 2.5” 10K 600GB SAS

‒ 2.5” 10K 900GB SAS

‒ 2.5” 15K 146GB SAS

‒ 2.5” Flash 200GB 3G and 6G SAS

‒ 3.5” Flash 400GB 3G and 6G SAS

‒ 3.5” SATA 2TB / 7200 RPM

‒ 3.5” 2TB Nearline SAS / 7200 RPM

New Drive Technology:

‒ 2.5” 15K 300GB SAS

‒ 2.5” Flash MLC 200GB 6G SAS

‒ 2.5” Flash MLC 400GB 6G SAS

‒ 3.5” 3TB Nearline SAS / 7200 RPM

NEW

RAID Group 2

RAID Group 1

RAID Group 3

Encryption Support Updates

All Virtual Storage Platforms are

encryption capable

‒ Every BED has an encryption capability built in

‒ Needs to be enabled by software license key

New encryption mode of

operations

‒ XTS-AES 256 bit encryption

Expanded key support

‒ 32 keys per array

‒ Encryption as access control

Other Encryption Benefits

Controller-based encryption for at-rest data

Targeted for customers who require confidentiality protections for data residing on media (not intended to be the primary confidentiality measure)

Encryption is performed in hardware with no degradation in throughput

Any media type internal to Hitachi Virtual Storage Platform is supported

Simplified and integrated key management; no external key management required

Green encryption

Hitachi Mainframe Storage Strategy

Leverage IBM Compatibility with Hitachi Value Add

Storage Subsystems

FICON, zHPF, FCP, MA, PAV, HyperPAV Hardware Interface

IBM Portfolio:

RCMF

Hyperswap Manager

GDPS

Hitachi Portfolio:

BC Manager,

Universal Replicator,

FDM, FC UVM, HDP

zOS + DFSMSdfp, Linux for System z, zVM, zVSE

IMS, CICS, DB2, MQSeries etc.

IBM

Storage

Hitachi

Storage

Hitachi

Storage

Solution

Level

Operating Systems

Middleware and

Application Level

Concurrent Copy, Flashcopy, Metro Mirror (PPRC),

Global Copy (PPRC-XD), z/OS Global Mirror (XRC) Storage Functions

IBM Compatibility

Hitachi

Strategy:

IBM Compatibility

IBM Compatibility

Development Focus

on Hitachi added

Value Solutions

(without Host SW)

(without Host SW)

(without Host SW)

Mainframe Vision and Direction

- Hitachi is bringing storage virtualization technology to the mainframe space

- HDP for Mainframe provides FCSE/DVE/EAV compatibility

- HDP is technology foundation for the next wave of mainframe virtualization

UVM SATA SAS SSD

SAS UVM

HDP Pool

HDP Pool + HDT

-Utilize existing assets

or low-cost external

storage

-Better space efficiency

-I/O load balancing

-Flexible volume

allocation

-Automatic Dynamic

Tiering

-Effective use of high-

speed SSD

EAV

DVE

EAV

DVE

EAV

DVE

FCSE EAV

DVE

EAV

DVE

EAV

DVE

EAV

DVE

EAV

DVE

EAV

DVE

FCSE

UVM

AGENDA






Summary

Virtual Storage Platform (VSP)

or

Storage Virtualization for Mainframe

Storage Virtualization for Mainframe

Constantly adding new devices to the Virtualization Support Matrix.

EMC VMAX added since August

IBM DS8800 added since August

Other storage vendor products added:

‒ XIOTECH

‒ 3PAR – V400 and V800

‒ Fujitsu Eternus

‒ DX60

‒ DX80

‒ DX90

‒ DX400

NEW

AGENDA






Summary

Provides fully compatible support for IBM features:

‒ EAV

‒ DVE

‒ FlashCopy SE

‒ FlashCopy SE – Performance Enhancements

‒ Change from Sync to A-Sync

Dynamic Provisioning

‒ Space Savings or “Thin Provisioning” characteristics

‒ Ultra Wide-Striping

‒ Physical pools comprised of multiple array groups

‒ 38MB pages striped across all storage in a pool

‒ Improved performance – especially for sequential workloads

NEW

Hitachi Dynamic Provisioning (HDP) for Mainframe

EXTENDED ADDRESS VOLUME (EAV)

z/OS 1.10 – IBM introduced Extended Address Volumes (EAV)

‒ Volumes

‒ Larger than 65,520 cylinders and 54GB

‒ Architectural limit raised to 268,434,453 cylinders

‒ Current implementation supports 262,668 cylinders

‒ Addresses the problem of z/OS® 4-digit UCBs

‒ Limit of 64K Unit Control Blocks (UCBs)

‒ Running out of addressable storage

‒ 3390-A volumes

‒ Maintains 3390 track format

‒ HyperPAV facilitates multiple accesses to a volume

‒ Addresses problem of queuing on a very large volume with a single UCB

‒ Track-managed and cylinder-managed space

EXTENDED ADDRESS VOLUME (EAV)

EAV DETAILS

Track-managed space

‒ Area on an EAV located within the 1st 65,520 cylinders

‒ Intended as storage for “small” data sets

‒ Extended Addressing Space (EAS) ineligible data sets

‒ Comparable to the same space on non-EAVs

‒ Space is allocated in track or cylinder increments

Cylinder-managed space

‒ Area on an EAV located above the 1st 65,520 cylinders

‒ Intended as storage for “large” EAS eligible data sets

‒ Space is allocated in multicylinder units (MCU) of 21 cylinders.

‒ System may round space requests up to a multiple of an MCU (i.e., 315 tracks)

DYNAMIC VOLUME EXPANSION (DVE)

Dynamic Volume Expansion Considerations

Enables transparent capacity increase to meet application demands ‒ Increases volume capacity for defined online volumes

‒ Pre-z/OS V1R10 up to 65,520 cylinders online ‒ z/OS V1R10 online expand past 65,520 cylinders online (3390-A)

Supports increased capacity with existing UCBs

‒ Provides easier management

‒ Reduces need to define new volumes requiring additional UCBs

‒ Does not require updating IOCD or sysgen for new device addresses

IBM® DFSMS device support code will detect the volume size change.

‒ Refreshes Volume Table of Contents (VTOC)

‒ Issues message IEA019I

‒ Creates System Management Facility (SMF) record type 22 with the old and new volume size

DVE cannot be used with volumes involved in replication relationships.*

- Relationship must be broken before expanding volume.

Note: *IBM restriction on DVE volumes in replication relationships

HITACHI DYNAMIC PROVISIONING FOR MAINFRAME BASICS

HITACHI DYNAMIC PROVISIONING FOR MAINFRAME

Optimize storage performance by spreading the I/O across all available disk drives

Optimize storage capacity by only allocating capacity that is actually used

Achieve FlashCopy space efficiency (FC-SE) for target volumes

Leverage DVE to seamlessly grow physical and logical capacity

LDEV LDEV LDEV LDEV LDEV LDEV LDEV LDEV

HDP Pool

HDP Volume

(Virtual 3390)

LDEVs

Mainframe Servers

Simplifies capacity expansion: no IOGEN; no documentation updates;

no changes to GDPS parameters, XRC sessions, FlashCopy configs, etc.

V-VOL

HOST

A B C

SHARED MEMORY

POOL-VOL

RAID group

SVP STORAGE

NAVIGATOR

Mapping management information

backup area

C A B

Hitachi Storage Navigator 2 Settings and reference of Dynamic Provisioning

are generally performed with Storage Navigator 2

Mapping management information Management area for managing mapping

information of V-VOL, pool, and Pool-VOL

and monitoring free capacity

Pool

Area created from 1 or more

RAID Groups and in which Pool-

VOLs are registered

V-VOL=DP-VOL (3390-A) Volumes that are accessed by Host

and have a capacity recognized by

Host but do not have physical area

Can be created with a larger

capacity than the installed physical

Pool-VOL

Pool-VOL (3390-V) Volume in which data is stored

Necessary to have enough physical

capacity for creating a Pool-VOL

HITACHI DYNAMIC PROVISIONING FOR MAINFRAME STORAGE

3390-A 3390-A

3390-A

PORT PORT PORT


DP

POOL

PAGE SIZE • 38MB • 672 3390 tracks

CAPACITY • Min: 8GB

• Max: 696TB (internal)

• Max: 928TB (external)

•1024 x 3390-V

•1024 x 3390-V

VOLUME EMULATION

TYPE

• 3390-V • Internal: 44 to 837,760 cyls

• External: 44 to 1,117,760 cyls

MINIMUM POOL SIZE • 9,676 cylinders • 8GB

DP-

VOLS

EMULATION TYPE • 3390-A • EAV volumes

CAPACITY • Minimum size 870KB

• Maximum size 218GB

• 1 cylinder

• 262,668 cylinders

3390-A VOLS > 65,520

CYLINDERS

• Rounded up to multiple of

1,113 cylinders

• 1,113 cyls = cylinder group of

25 mainframe HDP pages

RECOMMENDED SIZE • Multiple of 1,113 cylinders

USER 3390-A

DEFINITION

• 1-cylinder units • Not rounded up

WHEN DYNAMIC PROVISIONING IS USED

Virtual volume can be expanded using DVE without stopping host I/O

Volume redefinition is unnecessary

Initial: Purchased capacity 3GB

(V-VOL capacity 2GB)

Lack of free pool capacity due

to increase in used capacity

(1GB3GB)

Pool capacity expansion 3GB 6GB

Used

capacity 1GB

Used

capacity

3GB

V-VOL capacity seen from host

Virtual

volume

Purchased physical capacity

V-VOL

capacity

expansion

When expanding capacity, volume definition is unnecessary

(DP-VOL or pool capacity can be expanded without stopping host I/O).

Used

capacity

3GB

Pool capacity

expansion


SIMPLIFIED VOLUME DEFINITION, EXPANSION AND OPERATION

WHEN DYNAMIC PROVISIONING IS NOT USED

One RAID group has heavy workload, and it may become a performance bottleneck.

Resolution: Manually create a configuration that balances workloads on RAID groups

RAID Group 1 RAID Group 2

LDEV LDEV LDEV LDEV LDEV

RAID Group 1 has heavy workloads, and it

may become a performance bottleneck.


SIMPLIFIED DESIGN FOR PERFORMANCE LEVELING

WHEN DYNAMIC PROVISIONING IS USED

Actual capacity of the pool is created from multiple RAID groups

Volume allocated from actual capacity of pool and is spread across multiple RAID groups

- Distributes workloads across RAID groups

- Reduces the possibility that a specific RAID group becomes a performance bottleneck

RAID Group 1 RAID Group 2

Workloads are distributed to RAID

Group 1 and 2, and they are balanced.

V-VOL V-VOL V-VOL V-VOL V-VOL

Actual capacity pool


SIMPLIFIED DESIGN FOR PERFORMANCE LEVELING

DFSMS enables fully automated and highly efficient storage management.

‒ Automatic class selection (ACS) routines provide automation, which drives very high-capacity utilization

‒ Naming conventions (data class = DC)

‒ Backup and migration policies (management class = MC)

‒ Pooling (storage class = SC and storage group = SG)

Mainframe host SG view and HDP pool view of free space are not aligned.

‒ Updates and deletes to catalog, VTOC, VTOC-IX are not reflected in the DP-Pool.

Value of HDP for Mainframe

‒ Wide striping providing improved application performance

‒ Foundation for compatibility with EAV, DVE and FCSE

‒ Prerequisite for Hitachi Dynamic Tiering for Mainframe

April May

50%

80%

Space Utilization Host SG View

April May

50%

80%

Space Utilization DP-Pool View

MAINFRAME STORAGE MANAGEMENT CONSIDERATIONS

HITACHI DYNAMIC PROVISIONING FOR MAINFRAME STORAGE OVERPROVISIONING CONSIDERATIONS

Overprovisioning – NOT recommended for mainframe environments

‒ If used, customer must have a way to guarantee NEVER to run out of space

‒ Option 1

‒ Use DVE: IBM limitations on DVE require that with replication products an initial copy is required when volume size changes.

‒ Option 2

‒ Allocate DUMMY datasets to cover the overprovisioned space that is physically not available in the pool

INITIAL POOL SPACE

Total UCB capacity = 30TB

Initial pool space = 10TB

Dummy DSN space = 20TB

Pool space increase: Plus 10TB

Dummy space decrease: Minus 10TB

(Dummy must occupy same cylinders)

EXPANDED

POOL SPACE

DUMMY DATASETS

PHYSICAL

VIRTUAL

30T

B P

RO

VIS

ION

ED

CA

PA

CIT

Y

DUMMY DATASETS

DSN = data set name

SPACE RECLAMATION

HITACHI DYNAMIC PROVISIONING FOR MAINFRAME STORAGE SPACE RECLAMATION

Mainframe storage

‒ Uses complex count key data (CKD) structure and cylinder-head-record (CCHHR) addressing

‒ CKD-formatted storage contains control information even when there is no data.

‒ Hitachi Virtual Storage Platform monitors existence of user records on track and maintains status in track metadata.

Recovering space and restoring it to the Dynamic Provisioning pool when it is no longer used, making it available for use by other volumes

‒ If no user records exist on track, track is eligible for reclamation.

‒ All tracks in DP page must not have any user records.

Host initiated reclaim (HIR) for mainframe storage

‒ Reclaims pages that only have tracks with no user records

‒ Tracks metadata used to determine if user records exist on track

‒ Done during CCW execution

Mainframe storage with count key data (CKD) format and control

data may not contain zeros even after datasets have been deleted.

‒ Must have techniques to erase records and update metadata to indicate that track contains no user records and is eligible for reclamation

Techniques to make mainframe storage eligible to be reclaimed

‒ ICKDSF TRKFMT

‒ Resource Access Control Facility (RACF) with erase-on-scratch option enabled (Overhead will elongate Batch Elapsed Times)

‒ Other CCW sequences that follow our host initiated reclaim rules will also reclaim capacity with far less overhead.

HITACHI DYNAMIC PROVISIONING FOR MAINFRAME STORAGE SPACE RECLAMATION

INTEROPERABILITY WITH OTHER PROGRAM PRODUCTS

DYNAMIC PROVISIONING FOR MAINFRAME EAV AND DVE INTEROPERABILITY

Notes:

1. Any combination of 3390-A fixed and 3390-A DP-VOLs

2: To use 3390-A emulation with volumes smaller than 262,668 (max) cylinders, select desired size among 3390-3/9/L/M equivalent size. This does not require the CVS PP.

SUPPORTED 1 SUPPORT PLANNED

Customized Volume Size (CVS) 2

Hitachi Compatible PAV for Mainframe (HCPM)

Concurrent Copy XRC®

Hitachi TrueCopy®

Hitachi Universal Replicator (HUR) 2

FlashCopy® v2 / SE Hitachi ShadowImage® Hitachi Volume Migration Hitachi Universal Volume Manager

(UVM)

Dynamic Cache Residency

LDEV Security

LDEV Guard

NO SUPPORT PLANNED

FICON Data Migration (FDM)

Hitachi Cross-OS File Exchange

REPLICATION PRODUCTS – 3390 INTEROPERABILITY

Supported

Copy

Combinations

TrueCopy, Universal Replicator, XRC

P-VOL S-VOL

ShadowImage,

FlashCopy, VM,

S-VOL T-VOL

TrueCopy

HUR XRC SI FC VM VSP

VSP

USP,

USP V

VSP

3390-3/9/L/M

3390-A (Fixed)

Yes Yes No Yes No Yes No

3390-3/9/L/M

3390-A (DP-Vol)

Yes Yes No Yes No Yes Planned

3390-A (Fixed)

3390-3/9/L/M

No No No Yes No Yes No

3390-A (DP-Vol)

3390-3/9/L/M

No No No Yes No Yes Planned

HDP FOR MAINFRAME MIGRATION CONSIDERATIONS — SUMMARY

Three types of mainframe volumes:

1. Traditional 3390 –1,3,9,27,54 volumes – USP, USP V and VSP

2. Fixed size EAV volumes – 3390-A - VSP with V02+1 microcode

3. DVE capable EAV volumes – 3390-A DP-VOLs – VSP with V02+1 microcode

HDP for Mainframe supports interoperability between 3390-A volumes

‒ complete interoperability between 3390-A volumes with Hitachi ShadowImage, Hitachi TrueCopy, Hitachi Universal Replicator, etc.

‒ Limited 3DC scalability – currently only 1x1x1 (MxNxO planned in V3+1)

‒ Hitachi TrueCopy and Hitachi Compatible Mirroring for IBM® FlashCopy® supports interoperability between some combinations of all 3 emulation types

Migration options from traditional 3390-3/9/L/M to 3390-A DP Volumes

‒ Host-based migration using FDR/PAS or DFDSS, etc. (supported!)

‒ Use TrueCopy or XRC to migrate from USP or USP V to VSP, or VSP to VSP

AGENDA






Summary

Hitachi Dynamic Tiering for Mainframe

Dynamic Tiering

‒ Open Systems available since September 2010

‒ Mainframe available as of March, 2012

‒ Automated performance management

‒ Cost effective use of storage tiers

‒ Internal and external (virtualized) storage

‒ Managed in pages (42MB for OPEN, 38MB for Mainframe)

Cost savings through:

‒ Improved storage utilization

‒ Reduced storage requirements for Tier 1

‒ Less time spent on performance management

‒ Re-deploy personnel

‒ Integration with SMS

Mainframe Tiered Pools Dynamically Managed

Next step is to simplify storage administration by automating and eliminating the complexities of efficient use of tiered storage

Automatically move data on pages within HDP 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.

HDP Pool

Frequent Accesses

Infrequent References

Dynamic

Provisioning

Virtual

Volume

TIER 1

TIER 2

TIER 3

SSD

SAS

SLOW SAS

SAS

Slow SAS

EFD/SSD

TIER 1

TIER 2

TIER 0

POOL A

Last Referenced

Last Referenced

Different tiers of storage are now in

one pool of pages

Data is written to the highest

performance tier first

As it becomes less active it

migrates to lower level tiers

If activity increases it will be

promoted back to a higher tier

Since 20% of data accounts for

80% of the activity, only the active

part of a volume will reside on the

higher performance tiers.

Hitachi Dynamic Tiering: Page Level Tiering

Tier 0

Tier 1

Tier 2

0

IOP

S

Each DPVOL is analyzed

and relocated in turn

Page

IOPS break point where

pages higher than this

level are relocated to T0

IOPS break point where

pages destined for T1

and T2 is determined

Cycle

Page relocations

Monitor physical

IO to pages

Storage Resources as Tiers – Relocating Data based on IOPS

Subsystem

DFSMS Storage Groups and MF HDP and HDT

Tiered Storage can be implemented with DFSMS Storage Groups

DFSMS Attributes such as Data Class and Storage Class are used

to direct allocation to the right Tier

SSD Tier 1

SAS Tier 2

SATA Tier 3

HDT

Storage Group

SG2

SG3

SG1

Subsystem

ONE SG

• With MF HDP may combine all SG„s

into one and use HDT to move data

to the right Tier automatically based

on Workload performance

HDP Volumes

are managed

at page-level

Subsystem

Subsystem

Subsystem

DFSMS Storage Groups and MF HDT

HDT can be applied to selected Storage Groups only

DFSMS Horizontal Storage Groups and HDT Pools with

Dynamic Page based Tiering can be aligned

SSD Tier 1

FC/SAS Tier 2

SATA Tier 3

HDT

Storage Group

Online SG

Batch SG

Archive SG

Subsystem

Subsystem

Subsystem

DFSMS Storage Groups and MF HDT

HDT can be applied to selected Storage Subsystems only

DFSMS Vertical Storage Groups and HDP Pools with Dynamic Page

based Tiering can be aligned

SSD Tier 1

SAS Tier 2

SATA Tier 3

HDT

Storage

Group

Online SG

Batch SG Archive SG

AGENDA



Dynamic Provisioning – New for the Mainframe Environment



Summary

VSP – Other NEW Features

An Entire page of new features – some OPEN and some Mainframe

HDP, HDT, and Copy On Write (SNAP)

‒ Support for 5PB pool space (increased from 1.1PB

‒ Background scrubbing for FREE Pages

z/OS 1.13 - Full support

zTPF support

zHPF Multi-Track support

New RAID 6 implementation – 14+2

NDM – Non-Disruptive Migration (OPEN Systems ONLY)

HAM – High Availability Manager (OPEN Systems ONLY)

‒ Clustering of two or more VSP storage subsystems for automated fail-over

‒ Non-Disruptive fail-over and fail back.

NEW

Completely duplicated/

Interconnected recovery site

Remote disk mirroring

Local disk mirroring

Electronic vaulting

Tape onsite

Tape backup

offsite

More

Less Delayed

Immediate

More

Less Amount

of Data

Recovery

Time

Importance of

Data

Disk-to-disk backup and recovery

Remote PiT mediated copy

Out-of-region and multiple

data centre strategies

Data Protection Portfolio

Business Continuity – Mainframe Portfolio

In-System Replication

Solutions

Remote Replication

Solutions

Replication

Management Solutions

In-System Replication

Solutions

Remote Replication

Solutions

Replication Management

Solutions

ShadowImage

For full volume clones of

business data with

consistency

(ATTIME split with HUR)

TrueCopy

Synchronous, consistent clones at

remote location up to 300km

(~180 miles)

Business Continuity Manager (BCM)

Replication management in z/OS

environments

IBM Compatible

FlashCopy/SE

Point-in-time volumes and

datasets via IBM command

set in z/OS environment

Universal Replicator

Any distance, unique use of „Pull‟

technology and Journal Volumes

to accommodate link outages or

interruptions

Replication Manager w/ Mainframe

BCM plus “Enterprise-wide” GUI

replication management and monitoring

for mainframe and open systems

IBM Compatible XRC

Asynchronous remote replication

via compatible with IBM XRC in

z/OS environment

IBM Basic HyperSwap, GDPS

HyperSwap certification and/or

integration

High-availability storage for z/OS

environments

Replication Assessment, Migration and Implementation Services

Best practice designs that ensure results and lower risks in implementation

Replication Score Card (Health Check) Services

Periodic check-ups that ensure continued solution value

Split SI Pairs SI

Primary Site

JNL-VOL UR

JNL Group JNL Group

JNL-VOL

UR P-VOL UR S-VOL SI P-VOL

SI S-VOL

Linkage between HUR and ShadowImage

Enables consistent split of SI Pairs without suspending HUR replication

Facilitates DR testing or data repurposing

Maintains constant and consistent DR readiness on HUR secondary copy

Secondary Site

ATTime Split – Reducing Risk

Three data center Replication Solutions Mainframe

TrueCopy S-VOL shared as HUR P-VOL in intermediate site

Primary volume is shared P-VOL for two remote systems

VSP

HUR

JNL Group

P-VOL

VSP

S-VOL P-VOL

JNL-VOL TrueCopy

(Sync)

3DC Cascade*

S-VOL

VSP

JNL-VOL

JNL Group

VSP

JNL-VOL

HUR

JNL-VOL

Journal Group

P-VOL

TrueCopy (Sync)

3DC Multi-Target

S-VOL

VSP

JNL-VOL

JNL Group

Optional Delta Resync

VSP

S-VOL S-VOL

VSP

JNL-VOL

JNL Group

Both support up to 12x12x12 controller configurations

Four Data center Replication Solutions Mainframe

VSP

JNL-VOL

HUR

JNL-VOL

Journal Group

P-VOL

S-VOL

VSP

JNL-VOL

JNL Group

HUR

S-VOL

VSP

JNL-VOL

JNL Group

TrueCopy (Sync)

S-VOL

VSP

JNL-VOL

JNL Group

2DC

HUR

3DC

Cascade

3DC

Multi-target

AGENDA



Dynamic Provisioning – New for the Mainframe Environment



Summary

Mainframe Virtualization Summary

Easy to implement

Take advantage of storage tiers

Take advantage of advanced features and functions of the VSP

Extend the useful life of existing assets

Provide Mainframe connectivity to lower cost Midrange storage platforms

Single management point for all Mainframe storage

Non-disruptively move data within a single VSP frame

Non-disruptively move data across

multi-vendor storage frames

Improved storage utilization

Improved Quality of Service (QoS)

Virtual Storage Platform Summary

Reliable Built on a unique heritage of market

leading innovations Includes advanced data protection and

high availability capabilities A proven services organization and

network of partners reduce migration risk

Dynamic Adapts easily to your changing business

and IT requirements Automates data tiering dynamically Scales in 3 architectural dimensions – a

capability no other vendor offers

Open Supports a wide variety of multivendor

storage, operating systems and data types

Supports heterogeneous server virtualization environments

Would you like to know more ?

Webtech: http://www.hds.com/webtech/

Live and recorded for later replay.

Some recent mainframe related:

‒ 10/19/11 How to Apply the Latest Advances in Mainframe Storage

‒ 10/26/11 Find the Facts and Lies about Replication in Mainframe Storage

‒ 11/2/11 Why Hitachi Virtual Storage Platform Does So Well in a Mainframe Environment

http://www.hds.com/webtech/

http://www.hds.com/webtech/archive-by-year/




QUESTIONS AND DISCUSSION

THANK YOU

virtual storage platform virtualization, dynamic ... gse session 20120314.pdf · 3d scaling...

Documents