IBM DS8000 and IBM z Systems Synergy

IBM DS8000 and IBM z Systems Synergy
© Copyright IBM Corporation 2016
Agenda
• Synergy that optimizes Performance
• I/O Performance
• Networking Performance
• Management and Configuration Synergy
• Synergy that addresses Disaster Recovery and High Availability
• Data Protection and Backup
© Copyright IBM Corporation 2016
1
Performance
© Copyright IBM Corporation 2016
2
I/O Performance
© Copyright IBM Corporation 2016
3
z/OS Response Time Components
Service Time
Response
Time
=
IOSQ Time
+
Pending
Time
CMR
Delay
+
Disconnect
Time
+
Connect
Time
Device
Busy
Response Time
• IOSQ Time is the time measured when an I/O request is being queued in the system by z/OS.
• Pending Time represents the time that an I/O request waits in the hardware
• Disconnect Time results from a read miss and the control unit must access the disk to obtain the data
record requested by the application
• Connect Time is the time that the Disk, Channel, and Processor are connected during data transfer
• Interrupt Delay Time: the time measured from when the I/O operation is complete to when the
operating system begins to process the status
© Copyright IBM Corporation 2016
4
Four Stages of an I/O Operation
Response Time
Queuing Time
IOSQ Time
Service Time
Pending Time
Disconnect Time
Device in use (UCB
busy in IOS)
Device reserved from
another system
Read cache miss
Sometimes as a result
of device reserve as
HyperPAV aliases are
not used when devices
transition to/from
reserve state
Command response
(CMR) delay
Synchronous remote
copy – write with
synchronous replication
System Assist
Processor (SAP)
overhead
Multiple allegiance
or PAV write
extent conflicts
Device busy (domain
extent conflict)
Sequential write
hits, rate is faster
than controller can
accept
High FICON or switch
port utilization
Control unit busy
© Copyright IBM Corporation 2016
Connect Time
Channel data and
protocol transfer
5
IBM Innovations to Reduce z Systems I/O Response Times
IOSQ Time
Parallel Access
Volumes
HyperPAV
Pending Time
Multiple Allegiance
Disconnect Time
Connect Time
Adaptive Multi-Stream
Pre-Fetching (AMP)
MIDAWs
Intelligent Write Caching
(IWC)
z Systems High
Performance FICON
(zHPF)
Sequential Adaptive
Replacement Cache
(SARC)
FICON Express 16 Gb
channel
zHPF List Prefetch
Optimizer
4 KB cache slot size
zHyperWrite
Easy Tier integration
with DB2
Integrated DS8000 functions and features to address response time components
(not all functions listed)
© Copyright IBM Corporation 2016
6
Parallel Access Volumes (PAV)
• Function: Parallel Access Volumes
• Description: PAVs allow simultaneous I/Os to be initiated against the same
device number within the same z/OS system
• Why it matters: Reduces or eliminates IOS Queuing (IOSQ) time component
of I/O response time
For more information, see IBM DS8880 Architecture and Implementation, SG24-8323
© Copyright IBM Corporation 2016
7
Parallel Access Volumes
Without
PAV
• Application “B” starts an I/O
operation and succeeds as no
other application has an
outstanding I/O against device
100 (which is the base
address in this example)
• Application “A” in same z/OS
system starts and I/O but gets
held up because device 100 is
already busy with an I/O from
application “B”
• Application “C” in another
z/OS system starts an I/O but
is held up because device 100
is busy because Application
“B” has control of the device
© Copyright IBM Corporation 2016
8
Parallel Access Volumes
With
PAV
Application “B” starts an I/O
operation and succeeds as no
other application has an
outstanding I/O against device
100 (which is the base address in
this example)
Application “A” in same z/OS
system starts an I/O and IOS
obtains an alias of 1FE for device
100 (alias assigned by IOS)
Application “C” starts an I/O
against device 100 and receives
an alias assignment of 1FF by
IOS
Alias assignment can be static or
dynamic based on client
configuration settings.
© Copyright IBM Corporation 2016
9
Parallel Access Volumes – Static versus Dynamic
• In the original implementation of Parallel Access Volumes, aliases can be
defined as “static” or “dynamic”
• Static aliases
• Static alias assignment remains constant and can never move to another base device no matter
how intense the workload activity is
• Every system accessing that base device/alias combination must maintain that static relationship
• Dynamic aliases
• Dynamic alias assignment is fluid and throughout the day the aliases may be reassigned to
different base devices as needed to support the ever changing I/O workload
• Every system accessing that base device/alias combination must agree when an alias is to be
reassigned to another base address
• Within the Sysplex, Workload Manager (WLM) coordinates reassignment of an alias to another
base device
• There is no WLM management of dynamic aliases from one sysplex to another sysplex.
© Copyright IBM Corporation 2016
10
Parallel Access Volumes – Dynamic Alias Reassignment
Workload Manager monitors I/O activity and coordinates reassignment of
aliases across systems within the Sysplex. In this example, I/O demand
against device 100 requires the movement of an additional alias from device
110 to device 100 to satisfy goals set by WLM.
© Copyright IBM Corporation 2016
11
HyperPAV
• Function: HyperPAV
• Description: HyperPAVs extend the capability of PAVs by removing the need
for z/OS Workload Manager to coordinate the movement of aliases
• Why it matters: HyperPAVs react to changes in workload quicker than
dynamic aliases, reduces the overall number of aliases required, and helps to
enable larger CKD volumes.
• It also complements WLM I/O Priority management. With HyperPAV, when an I/O
request finishes the next I/O to run for the device is the highest priority request for that
CU image. Prior to HyperPAV the highest priority request for that one device was the
next to run.
• A virtualization technology that effectively multiplies the number of device addresses
available as each OS image can use the same alias device address for a different base
at the same time.
For more information, see IBM DS8880 Architecture and Implementation, SG24-8323
© Copyright IBM Corporation 2016
12
HyperPAV
PRD1
PRD2
PRD3
TEST
Total
Dynamic
PAVs
12
8
10
2
32
HyperPAVs
12
12
12
12
12
© Copyright IBM Corporation 2016
•
With Dynamic PAVs, Workload
Manager controls PAV
reassignment across all
systems within the Sysplex
•
WLM only coordinates alias
movement every 10 seconds
which can result in some
sluggishness in alias
reassignment
•
With HyperPAVs, WLM no
longer controls reassignment
and each individual z/OS
system controls its own alias
assignment without the need
to coordinate cross systems
•
Aliases are used immediately
from the pool and returned to
the pool when no longer in use
•
HyperPAV generally reduces
the number of aliases required
13
Multiple Allegiance
• Function: Multiple Allegiance
• Description: Without multiple allegiance, an I/O from another system to a
shared disk device with an active I/O operation would be held.
• Why it matters: Multiple Allegiance reduces pending time by allowing
concurrent I/Os to proceed to shared devices when no domain conflict exists.
For more information, see IBM DS8880 Architecture and Implementation, SG24-8323
© Copyright IBM Corporation 2016
14
Multiple Allegiance
•
Before multiple
allegiance, if an
application in one
system had an I/O
started to a device,
another system could
not start an I/O to the
same device to avoid a
domain conflict. This
single allegiance
guaranteed that the
control unit had access
to the device for the
remainder of the
channel program
•
With multiple allegiance,
I/O requests are
accepted by the
DS8000 and processed
in parallel unless there
is a conflict when writing
to the same domain.
© Copyright IBM Corporation 2016
15
I/O Priority Queuing
• Function: I/O Priority Queuing
• Description: Permits z/OS Workload Manager to indicate a priority for I/Os.
• Why it matters: The DS8000 can process many concurrent I/Os. But when
channel programs conflict with each other and must be serialized to ensure data
consistency, I/O Priority Queuing will accept a priority from zWLM to process
higher importance work before lower importance work. This reduces contention
in a shared environment and reduces response times.
For more information, see IBM DS8880 Architecture and Implementation, SG24-8323
© Copyright IBM Corporation 2016
Do not confuse I/O Priority Queuing with I/O Priority Manager
16
I/O Priority Queuing
Lower priority
work will
increase in
importance
over time to
ensure the
lower priority
workload is
processed
© Copyright IBM Corporation 2016
Do not confuse I/O Priority Queuing with I/O Priority Manager
17
Modified Indirect Data Access Word (MIDAW)
• Function: Modified Indirect Data Access Word (MIDAW)
• Description: MIDAWs are a new method of gathering data into and scattering
data from discontiguous storage locations during an I/O operation. MIDAWs
benefit small record I/O.
• Why it matters: Enabling and using MIDAW reduces the FICON channel
utilization and reduces the channel connect time to transfer data blocks.
MIDAW relies on z/OS Media Manager and datasets enabled for Media
Manager include extended format, PDSE and VSAM datasets such as Linear.
DB2 is a heavy user of Linear VSAM and Extended Format datasets but there
are many other examples. By implementing MIDAW, you will reduce the
number of FICON channel frames and exchanges flowing across the FICON
channel.
For more information, see How Does the MIDAW Facility Improve the Performance of FICON, REDP-4201
© Copyright IBM Corporation 2016
18
High Performance FICON for IBM z Systems - zHPF
• Function: zHPF
• Description: zHPF is an enhanced FICON protocol and I/O architecture
designed to reduce protocol exchanges between the channel and the storage
control unit.
• Why it matters: Traditional FICON is a protocol that requires multiple
exchanges in order to transfer data. zHPF reduces the protocol exchanges to a
minimum to reduce channel utilization and reduce connect time associated with
channel activity. Advanced in hardware enables IBM to optimize hardware
assists while preserving all of the z QoS including end-to-end data integrity
checking which is transparent to applications and middleware, in-band I/O
measurements, workload management, etc.
For more information, see IBM DS8870 and IBM z Systems Synergy, REDP-5186
© Copyright IBM Corporation 2016
19
zHPF Overview
Without
zHPF
Example of 4KB read
Six protocol exchanges
using CCWs
© Copyright IBM Corporation 2016
With
zHPF
Example of 4KB read
Three protocol exchanges
using TCWs
20
zHPF Evolution
Version 1
•
•
•
•
Single domain, single
track I/O
Reads, update writes
Media Manager
exploitation
z/OS 1.8 and above
Version 2
•
•
•
Multi-track but <= 64K
Multi-track any size
Extended distance I
Version 3
•
•
•
•
•
© Copyright IBM Corporation 2016
Format writes
Multi-domain I/O
QSAM/BSAM/BPAM
exploitation
z/OS R1.11 and
above
EXCPVR
Version 4
•
•
•
•
EXCP Support
ISV Exploitation
Extended Distance
II
SDM, DFSORT,
z/TPF
21
zHPF and DB2 – Working Together
• DB2 functions are improved by zHPF
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
DB2 database reorganizations
DB2 incremental copy
DB2 LOAD and REBUILD
DB2 queries
DB2 RUNSTATS table sampling
Index scans
Index-to-data access
Log applies
New extent allocation during inserts
Reads from a non-partition index
Reads of large fragmented objects
Recover and restore functions
Sequential reads
Table scans
Write to shadow objects
© Copyright IBM Corporation 2016
22
DB2 list pre-fetch optimizer
• Function: DB2 list pre-fetch optimizer
• Description: Enhancement to zHPF providing improvement to DB2 list
prefetch processing when utilizing 4 KB pages.
• Why it matters: Specialized set of instructions in the read command allow the
control unit to accept and anticipate the next set of records to be read
(prefetched) while the host application is reading the current set of records.
Used for scattered I/O processing by DB2 to reduce disconnect time for
disorganized index scans, list prefetch access path and other DB2 uses.
For more information, see DB2 for z/OS and List Prefetch Optimizer, REDP-4862
© Copyright IBM Corporation 2016
23
DB2 list pre-fetch optimizer
• FICON – List prefetch channel
program requires two
commands for each DB2 page
read (locate record (seek) and
and buffer read
• zHPF – List prefetch sends list
of seek addresses and will read
32 DB2 pages using just three
or four commands
• Benefit is more efficient
channel programs, more
parallelism in I/O access,
prestage records for DB2 using
overlapping buffer processing
Beneficial for DB2 disorganized index scans
© Copyright IBM Corporation 2016
24
Extended Distance FICON I
• Function: Extended Distance FICON I
• Description: Extended Distance FICON is used in z/OS Global Mirror (XRC)
environments where the System Data Mover is separated by distance from the
primary DS8000 storage controller.
• Why it matters: When using z/OS Global Mirror, Extended Distance FICON
can help to reduce the need for additional channel extender software by
increasing the number of read commands in flight. The result of using Extended
Distance FICON maintains performance of FICON channels over very long
distances.
For more information, see IBM DS8870 and IBM z Systems Synergy, REDP-5186
© Copyright IBM Corporation 2016
25
Extended Distance FICON I
• Extended Distance FICON
extends the FICON
architecture by implementing
a new information unit (IU)
pacing protocol.
• Allows up to 255 IUs in flight
versus the standard 16.
• Allows simpler channel
extension hardware to be
used.
• Used by z/OS Global Mirror
System Data Mover read
record set (RRS) commands
• May eliminate the need to
have XRC Emulation running
on the channel extension
hardware
© Copyright IBM Corporation 2016
26
Extended Distance FICON II
• Function: Extended Distance FICON II
• Description: Reduces the impact of distance on I/O response times by 50% for
large data writes, providing significant response time improvements when
leveraging HyperSwap over Metro Mirror distance.
• Why it matters: Reduces the impact of distance on I/O response times by 50%
for large data writes, providing significant response time improvements when
leveraging HyperSwap at distance. Applications such as DB2 tend to use larger
sized log writes.
For more information, see IBM DS8870 and IBM z Systems Synergy, REDP-5186
© Copyright IBM Corporation 2016
27
HyperSwap without zHPF Extended Distance II
Pre-HyperSwap
Channel
zHPF
Secondary
DB2 Utilities
(256 KB or 512 KB
Write)
Interlocked
exchanges
Primary
Metro Mirror (FCP)
Metro Mirror pre-deposit write and
stream of tracks
Post-HyperSwap
zHPF
Channel
At least four interlocked exchanges
at long distance. At 10 km, ~400
usec added to each I/O (50%
penalty).
DB2 Utilities
(256K Write)
Secondary
Primary
© Copyright IBM Corporation 2016
Note: DB2 utility writes in
V11 moved to 512K, so the
disparity increases
28
HyperSwap with zHPF Extended Distance II
Channel
zHPF
Command
Data
Status
Primary
Pre-HyperSwap
Secondary
Metro Mirror (FCP)
Metro Mirror pre-deposit write
and stream of tracks
Post-HyperSwap
zHPF
Command
Data
Status
Channel
zHPF Extended Distance II
will execute most write
operations in one round trip
Secondary
Primary
© Copyright IBM Corporation 2016
29
DB2 Integration With Easy Tier
• Function: DB2 Integration With Easy Tier
• Description: The Easy Tier API allows DB2 to proactively instruct Easy Tier of
the intended usage of DB2 files
• Why it matters: DB2, using Media Manager, can query the DS8870/DS8880 to
determine dataset placement, thus eliminating the need for Easy Tier to learn
the file’s heat resulting in better database performance
For more information, see IBM DS8870 and IBM z Systems Synergy, REDP-5186
© Copyright IBM Corporation 2016
30
DB2 Integration With Easy Tier
• DB2 database reorganization under z/OS
• Without Easy Tier Application for System z
• Database extents may currently be on high performance tier
• Database reorganization will move parts of the database to new
extents
• To Easy Tier, these extents may have been cold since they were
previously free space on the volume
• This may result in database extents now being relocated to lower
performance tiers
• With Easy Tier Application for System z
• z/OS will query the DS8870/DS8880 to obtain average heat and tier
assignments
• During the database reorganization z/OS will specify the desired
target tier
© Copyright IBM Corporation 2016
31
DB2 Integration With Easy Tier
Without Easy Tier Application
for System z
It takes time for ET to optimize the target
datasets after re-org
With Easy Tier Application for
System z
Target datasets are already optimized before the
re-org completes
DB2
DB2
Re-org
Storage
Storage
© Copyright IBM Corporation 2016
Assignment
hint
Re-org
SSD
SSD
HDD
HDD
32
zHyperWrite
• Function: zHyperWrite
• Description: IBM zHyperWrite is a cooperative technology provided by
DS8870/DS8880 and exploited by z/OS with GDPS or TPC-R controlled
HyperSwap enabled environments
• Why it matters: Reduces critical DB2 log write latency by bypassing control
unit overhead while operating in synchronous Metro Mirror environments by
combining DS8870/DS8880 and software mirroring functions within Media
Manager resulting in reduced disconnect time
For more information, see IBM DS8870 and IBM z Systems Synergy, REDP-5186
© Copyright IBM Corporation 2016
33
Traditional DB2 Log Write with Metro Mirror
1.
2.
3.
4.
DB2 log write to Metro Mirror primary
Write mirrored to secondary
Write acknowledged to primary
Write acknowledged to DB2
© Copyright IBM Corporation 2016
34
zHyperWrite
• Improved DB2 Log Write
performance with DS8870/DS8880
Metro Mirror
• Reduces latency overhead compared
to normal storage based synchronous
mirroring
• Reduce write latency and
improved log throughput
• Up to 40% improvement in testing
© Copyright IBM Corporation 2016
35
DB2 Log Write with zHyperWrite
1. DB2 log write to Metro Mirror
primary and secondary in parallel
2. Writes acknowledged to DB2
In this case, Metro Mirror does not
mirror the log writes
© Copyright IBM Corporation 2016
36
System Optimized Caching
• Function: System optimized caching
• Description: Cache algorithms designed specifically for System z workload.
Optimized cache slot size of 4 KB to match System z
• Why it matters: Improves read cache hit rates and sequential pre-fetch to
reduce application I/O response times for both reads and writes. Smaller cache
slot size reduces cache waste when accessing random data.
For more information, see IBM DS8880 Architecture and Implementation, SG24-8323
© Copyright IBM Corporation 2016
37
System z Optimized Caching
• Sequential Prefetching in Adaptive Replacement Cache (SARC)
• Self learning algorithm
• Adaptively and dynamically learns what data should be stored in cache
• Dynamically divides the cache between random and sequential lists to maximize the throughput
obtained for sequential workloads
• Improves cache hits by up to 100% and improves I/O response time by 25%
• Adaptive Multi-Stream Pre-fetching (AMP)
• Dynamically decides what and when to pre-fetch
• Manages what is in the sequential cache list
• Two fold increase in sequential read throughput
• Intelligent Write Caching
• Better write cache management and destage order
• Up to two times throughput for random write workloads
© Copyright IBM Corporation 2016
38
DS8000 Cache Efficiency
Example of storing two random 4 KB data blocks in cache
Two 4K cache segments allocated (4K stored, 0K unused)
DS8000
blk1
4KB
slots
blk2
Two 4K cache segments allocated (8K stored, 24K unused)
VSP
blk1
16KB
slots
blk2
Unused space
Unused space
Two 4K cache segments allocated (8K stored, 120K unused)
VMAX
blk1
64KB
slots
blk2
Unused space
Unused space
VMAX requires 2n GB cache to support n GB of “usable” cache
© Copyright IBM Corporation 2016
39
IMS WADS and DS8870/DS8880 Integration
• Function: IMS Write-Ahead Data Set (WADS) and DS8870/DS8880
Integration
• Description: IMS enhancement which provides indication or hint to
DS8870/DS8880 that this is a WADS file write. DS8870/DS8880 ensures that
data is in cache and avoids a write miss as WADS are update type writes.
• Why it matters: IMS WADS writes are extremely time sensitive and are critical
log records used for recovery in conjunction with the IMS Online Log Data Set
(OLDS). Ensuring updated records are cache write hits reduces or eliminates
disconnect time for WADS I/Os improving transaction response times.
For more information, see IBM DS8870 and IBM z Systems Synergy, REDP-5186
© Copyright IBM Corporation 2016
40
IMS WADS and DS8870/DS8880 Integration
• IMS WADS is used for database logging
• Host software provides an indication to the DS8000 that this is WADS
• DS8000 ensures data is in cache and avoids a write miss
IMS
z/OS
WADS without enhancement
WADS with enhancement
Example measurement: This DS8000 enhancement reduced disconnect time of WADS IOs to 0
© Copyright IBM Corporation 2016
41
DS8000 I/O Priority Manager and Work Load Manager
• Function: I/O Priority Manager and zWLM Integration
• Description: I/O Priority Manager allows z/OS Workload Manager to set I/O
importance and I/O goal achievement history in I/O requests that are used by
the DS8000 to prioritize I/Os when resources become constrained
• Why it matters: Allows more effective storage consolidation while still
maintaining I/O performance levels for critical workloads even if storage
resources are constrained. Quality of Service (QoS) guarantees that priority
work will get serviced.
For more information, see DS8000 I/O Priority Manager, REDP-4760
© Copyright IBM Corporation 2016
42
I/O Priority Manager and Work Load Manager
• Application A and B initiate an I/O operation to the same DS8870/DS8880 rank (may be different
logical volumes)
• zWLM sets the I/O importance value according to the application priority as defined by system
administrator.
• If resources are constrained within the DS8870/DS8880 (very high utilization on the disk rank), I/O
Priority Manager will handle the highest priority I/O request first and may throttle low priority I/Os to
guarantee a certain service level
© Copyright IBM Corporation 2016
43
Performance - Networking
© Copyright IBM Corporation 2016
44
16 Gb/second FICON
• Function: 16 Gb/second FICON
• Description: DS8870/DS8880 provides 16 Gb/second FICON host adapters to
integrate in with the IBM z13 processor’s FICON Express 16S and SAN fabric
• Why it matters: 16 Gb/second FICON ports improves I/O rates and bandwidth
and reduces connect time transferring data in the channel. This is especially
evident when large block transfers are taking place
Check out the IBM z13, FICON Express 16S and the DS8870
on YouTube at https://www.youtube.com/watch?v=OGuzeSdnEp8
For more information, see Get More Out of Your I/T Infrastructure with IBM z13 I/O Enchancements, REDP-5134
© Copyright IBM Corporation 2016
45
16 Gb/second FICON
• DS8880 can have up to 32 16
Gb/second host adapters
• DS8870 can have up to sixteen (16) 16
Gb/second host adapters
• Negotiates to connect at 16 Gb / 8 Gb or 4 Gb
• Supports both FICON and FCP attachment
• Both Long Wave and Short Wave options
available
• Each adapter provides four ports
• Faster z System channels, FICON/FCP
switches and DS8000 ports reduces
connect time
© Copyright IBM Corporation 2016
46
Forward Error Correction (FEC)
• Function: Forward Error Correction (FEC)
• Description: New T11 standard that provides error correction on top of
64B/66B encoding and improves reliability by reducing bit errors
• Why it matters: Ensures data integrity, guarantees higher and more
deterministic performance and robust data security at extremely high data rates.
Enables faster network speeds over existing cabling infrastructures and reduces
retransmits and timeouts.
See forward error correction in action on YouTube at https://www.youtube.com/watch?v=UKfiFEyfIiY&feature=youtu.be
For more information, see Get More Out of Your I/T Infrastructure with IBM z13 I/O Enchancements, REDP-5134
© Copyright IBM Corporation 2016
47
Forward Error Correction (FEC)
64b/66b ECC
versus
8/b10b ECC
Corrects up to
11 bit errors in a
2112 byte block
• Allows FICON channels to
operate at higher speeds, over
longer distances with reduced
power and higher throughput
while retaining traditional RAS
levels
• Used for controlling errors in
data transmission over
unreliable or noisy
communications channels
• Sender encodes messages in
a redundant way by using
error correcting code (ECC)
which allows receiver to detect
a limited number of bit errors
that might occur anywhere in
the message and often
corrects these errors without
retransmission
© Copyright IBM Corporation 2016
48
Read Diagnostic Parameters (RDP)
• Function: Read Diagnostic Parameters (RDP)
• Description: New T11 Read Diagnostic Parameters Extended Link Service
(RDP ELS) defines a method for SAN management software in the fabric to
retrieve standard counters that describe the optical signal strength (send and
receive), error counters and other critical information for determining the quality
of the link
• Why it matters: Diagnostic parameters can assist in finding the root cause for
problematic links in the SAN which may be caused by cabling infrastructure,
patch panels, host adapters, server channels and optics. Reduces the time it
takes to identify problem fault and to isolate the failed or failing component.
For more information, see Get More Out of Your I/T Infrastructure with IBM z13 I/O Enchancements, REDP-5134
© Copyright IBM Corporation 2016
49
Read Diagnostic Parameters (RDP)
After a link error is detected (e.g. IFCC, CC3 reset event, link incident report), use link data
returned from Read Diagnostic Parameters to differentiate between errors due to failures in
the optics versus failures due to dirty or faulty links
showioport -metrics
© Copyright IBM Corporation 2016
50
FICON Dynamic Routing (FDR)
• Function: FICON Dynamic Routing (FDR)
• Description: Enables exploitation of SAN dynamic routing polices in the fabric
to lower cost and improve performance for supporting I/O devices
• Why it matters: Port based routing schemes traditionally assign static interswitch links (ISLs) at fabric login time via a round robin allocation algorithm
which may cause some ISLs to become overloaded while others are
underutilized. FICON Dynamic Routing enables sharing of ISLs for FICON
traffic and FCP Metro Mirror traffic. Provides for better utilization of available
ISL bandwidth and simplifies management of inter-switch links
For more information, see Get More Out of Your I/T Infrastructure with IBM z13 I/O Enchancements, REDP-5134
© Copyright IBM Corporation 2016
51
FICON Dynamic Routing (FDR)
Static Routing Policy
(Using Brocade Port Based Routing)
Port 6
Port 7
Switch 10
Switch 20
Port 8
ISL assignment is static until next fabric login based on first come - first served. ISL may be under or
over utilized depending on I/O activity to the devices assigned to that ISL.
IODF has distant switch and port – route over ISL is determined at fabric login
© Copyright IBM Corporation 2016
52
FICON Dynamic Routing (FDR)
Static Routing Policy
(Using Brocade Device Based Routing and Cisco default static routing)
Port 6
Port 7
Switch 10
© Copyright IBM Corporation 2016
Switch 20
Port 8
53
FICON Dynamic Routing (FDR)
Dynamic Routing Policy
(Using Brocade Exchange Based routing and Cisco Open Exchange ID routing)
Switch 10
Switch 20
Every I/O’s path is determined at execution time consistent with ISL utilization
© Copyright IBM Corporation 2016
54
Management and Configuration
© Copyright IBM Corporation 2016
55
Extended Address Volume (EAV)
• Function: Extended Address Volumes (EAV)
• Description: Allows CKD (3390) devices sized up to 1 TB (1,182,006
cylinders) – far beyond the original “large” 3390-9 model type
• Why it matters: Many large z System environments have thousands of
devices and even with channel subsets, are exceeding the per image maximum
of 64K devices. Furthermore, fewer larger volumes makes management easier
for volume management tasks such as backups, SMS allocation, etc.
For more information, see IBM DS8880 Architecture and Implementation, SG24-8323
© Copyright IBM Corporation 2016
56
Extended Address Volume (EAV)
• Continued exploitation by z/OS
•
•
•
•
•
•
Non-VSAM extended format datasets
Sequential datasets
PDS
PDSE
BDAM
BCS/VVDS
• Large volumes to reduce
management efforts
• Create EAV dynamically with
dynamic volume expansion from
smaller to larger volumes
• Up to 1,182,006 cylinders in size
(1 TB) versus old limit of 65,520
cylinders
Cylinder Region
Track Region
Track managed region uses 16 bit cylinder and 16 bit track address (CCCCHHHH)
Cylinder managed region uses 28 bit cylinder and 4 bit track address (CCCCCCCH)
© Copyright IBM Corporation 2016
57
Dynamic Volume Expansion
• Function: Dynamic Volume Expansion
• Description: Allows a 3390 CKD volume to be expanded dynamically
concurrent with normal I/O activity
• Why it matters: Simplifies management by enabling easier online volume
expansion for z Systems. Dynamically modify a volume size to increase in size
up to 1 TB while keeping volume online.
For more information, see IBM DS8870 and IBM z Systems Synergy, REDP-5186
© Copyright IBM Corporation 2016
58
Dynamic Volume Expansion
• Dynamic volume expansion
available for any CKD volume
• Simplifies management effort and
enables growth
Dynamically increase the size of a CKD
volume while online and active I/O
• Any 3390 can be increased in size
up to the maximum size
• On z/OS systems, if a volume is
dynamically expanded, the VTOC
and VTOC index can be
automatically reformatted to map the
additional space (enabled via
SYS1.PARMLIB(DEVSUPxx) member
REFVTOC=ENABLE) (If additional space is
required in the VTOC then a manual ICKDSF is
required to extend the VTOC size.)
© Copyright IBM Corporation 2016
59
Quick Initialization
• Function: Quick Initialization
• Description: DS8000 makes new CKD logical volumes immediately available
and accessible to the host as soon as it is created.
• Why it matters: New volumes, and volumes resized via dynamic volume
expansion, can be immediately read and written to by the host applications
without the need to wait for formatting by the DS8000. Copy Services functions
can also be established as soon as the device is created.
For more information, see IBM DS8870 and IBM z Systems Synergy, REDP-5186
© Copyright IBM Corporation 2016
60
Volume Formatting Overwrite Protection
• Function: Volume Formatting Overwrite Protection
• Description: DS8000 Query Host Access function enables systems to
determine if a device (volume) is online to any system.
• Why it matters: By integration with IBM Device Support Facility (ICKDSF)
utility, the DS8000 can inform the software trying to initialize or reformat a
volume is that volume is online to any other system and prevent accidental data
loss.
For more information, see IBM DS8870 and IBM z Systems Synergy, REDP-5186
© Copyright IBM Corporation 2016
61
Volume Formatting Overwrite Protection
• Standard z System utility, ICKDSF is
integrated with DS8000 to determine if a
device is online
• ICKDSF parameter VERIFYOFFLINE will
fail an initialize or reformat task if the
device is online to any other system or
sysplex
1. LPAR “B” wants to initialize device
1000 on DS8000.
2. Program ICKDSF checks
online/offline status and determines
that LPAR “A” has device online
3. ICKDSF fails because
VERIFYOFFLINE queries the device
status and sees LPAR “A” currently
has device online
© Copyright IBM Corporation 2016
62
Query Host Access
• Function: Query Host Access
• Description: Used to validate that replication secondary and FlashCopy target
volumes are offline
• Why it matters: In large multi-system environments, it is often difficult to
ascertain which system potentially has a volume online. Replication, such as
Metro Mirror and volume level FlashCopy require that the secondary or target
device be offline to all systems. Query Host Access provides the means to
query and identify which systems are using the selected device.
For more information, see IBM DS8870 and IBM z Systems Synergy, REDP-5186
© Copyright IBM Corporation 2016
63
Control Unit Initiated Reconfiguration (CUIR)
• Function: Control Unit Initiated Reconfiguration
• Description: CUIR varies a path or paths offline to all z Systems hosts to allow
service to an I/O enclosure or host adapter, then varies on the paths to all host
systems when the host adapter ports are available. This function automates
channel path management in z Systems environments in support of selected
DS8000 service actions.
• Why it matters: z System connectivity can be complex and even more
complex if multiple LPARs or systems are sharing common ports. CUIR
reduces manual operator intervention and the possibility of human error during
maintenance actions. CUIR also reduces the time that is required for the
maintenance window.
For more information, see IBM DS8880 Architecture and Implementation, SG24-8323
© Copyright IBM Corporation 2016
64
Control Unit Initiated Reconfiguration
• As part of installation, SSR enables or
disables CUIR as part of configuration
option settings based on customization
worksheets provided by client (from
DS8000 Introduction and Planning Guide
appendix)
• If service event requires taking a path or
host adapter offline, SSR can utilize
CUIR to take affected paths offline to
z/OS or z/VM systems
• If CUIR can not take a path offline, the
DS8000 is notified
• Eliminates customer operations from
having to determine which paths to vary
offline/online and possibly shorten
service event.
• CUIR can be enabled/disabled by SSR
at any time
• CUIR does not have to be used if
enabled
© Copyright IBM Corporation 2016
CUIR is enabled by the SSR based on the configuration worksheets
Which cable
was that
again?
65
IOSSPOFD Tool
• Function: A tool for discovering single points of failure in an I/O configuration
• Description: IOSSPOFD can be used to determine if volumes have a singular
hardware component common to all paths to the device. It can also be used to
determine if a primary volume and an alternate volume have any common
hardware components. If such a hardware component were to fail then the
volume would not be available.
• Why it matters: Complex I/O configurations can be difficult to document and
maintain. New servers, SAN hardware and disk systems are continually being
upgraded and replaced. The potential to have a single point of failure in the
design of the I/O configuration is high.
For more information, see z/OS Unix Tools, Tools and Toys
© Copyright IBM Corporation 2016
66
IOSSPOFD Tool
• IOSSPOFD tool can run to check for single points of failure
• Run as batch job or called from a program, output goes to the job log
• List of devices, volumes or data sets
• Compare pairs of devices, volumes or data sets
• Available at z/OS tools and toys website
• http://www-03.ibm.com/systems/z/os/zos/features/unix/bpxa1ty2.html
Checking individual devices for single points of failure
DEVLIST(410,411,980-9A0)
VOLLIST(SYSRES,WORK*,TEST01)
DSNLIST(SYS1.NUCLEUS,SYS1.LINKLIB,DB2.DATABASE)
Checking pairs of devices for single points of failure between them
DEVN1(0410) DEVN2(1410)
VOLSER1(RACFPM) VOLSER2(RACFAL)
DSN1(SYS1.RACF.PRIMARY) DSN2(SYS1.RACF.ALT)
IND_CHECKS(YES|NO)
© Copyright IBM Corporation 2016
67
Enhanced Single Point of Failure Messages
• Control unit related single points of failure (OA48188)
• Specific messages will describe the single points of failure
• Odd/even enclosure and rack will no longer be considered SPOFs
• Requires DS8000 7.5 level microcode, IOS may support for older levels/models
Old
New
IOSPF209I Volume PLX06A (5D3F) has all control unit interfaces share 1 of 4 common components.
IOSPF209I Volume PLX06A (5D3F) – all control unit interfaces share the following hardware
components: Enclosure, host adapter card
• Channel related single points of failure are described (OA47210)
Old
New
IOSPF254I Volumes SACP01 (1002) and SBCP01 (2002) have all CHPIDs share a single point of
failure, component indicators = 10000000_00000000_00000000 (model dependent).
IOSPF254I Volumes SACP01 (1002) and SBCP01 (2002) - all CHPIDs share the following hardware
components: Processor book, fanout card, I/O domain, channel card
© Copyright IBM Corporation 2016
68
Disaster Recovery and High Availability
© Copyright IBM Corporation 2016
69
FlashCopy Software Integration
• Function: FlashCopy Software Integration
• Description: Specific FlashCopy interfaces for z Systems environments
• Why it matters: Tool set for invoking FlashCopy integrated into standard z
System utilities eliminate the need for special interfaces, exits and testing.
For more information, see IBM DS8870 Copy Services for IBM z Systems, SG24-6787
© Copyright IBM Corporation 2016
70
FlashCopy Software Integration
Option
DFSMSdss
TSO/Rexx
ICKDSF
z/VM CP
z/VSE IXFP
SNAP
Volume
FlashCopy
Dataset
FlashCopy
Space-Efficient
FlashCopy
Remote Pair
FlashCopy
FlashCopy
“nocopy”
Consistency
Group
FlashCopy
Incremental
FlashCopy
Display a list of
FlashCopy
relationships
Volume Defrag
© Copyright IBM Corporation 2016
71
Remote Mirror Software Integration
• Function: Remote Mirror Software Integration
• Description: Specific remote copy interfaces for z Systems environments
supporting Metro Mirror, Global Copy, Global Mirror and Metro/Global Mirror
• Why it matters: Tool set for invoking peer-to-peer copy functions integrated
into standard z System utilities eliminate the need for special interfaces, exits
and testing.
For more information, see IBM DS8870 Copy Services for IBM z Systems, SG24-6787
© Copyright IBM Corporation 2016
72
Remote Mirror Software Integration
Option
z/OS
TSO/
Rexx
z/OS
ICKDSF
z/OS
ANTRQST
API
GDPS
z/VM
ICKDSF
z/VSE
ICKDSF
z/TPF
z/TPF
ICKDSF
Path
management
Freeze/Thaw
Copy Pair
management
Recovery
secondary
Suspend/resume
Failover/failback
© Copyright IBM Corporation 2016
73
z/OS HyperSwap
• Function: z/OS HyperSwap
• Description: Provides a high availability solution in the event of a failure in a
control unit, FICON network or other permanent errors which prevents
continued operation to a primary disk subsystem. HyperSwap is an integral part
of z/OS and is fully enabled by utilizing GDPS or TPC-R
• Why it matters: A dynamic switch from primary to secondary disk while
maintaining continued operations during the switch helps to alleviate downtime
caused by hardware failures. Fully automated switchover facilitated by
GDPS/HyperSwap or TPC-R HyperSwap capabilities.
For more information, see IBM DS8870 and IBM z Systems Synergy, REDP-5186
© Copyright IBM Corporation 2016
74
HyperSwap
• HyperSwap provides high availability by
substituting Metro Mirror secondaries for Metro
Mirror primary devices transparently
• Can swap large number of devices fast and
nondisruptively
• Changes status in secondary disk subsystem
• Transparent to applications
application
Defined
disk #1
Defined
disk #2
• Swap
• Mirror is suspended and Changed Recording bitmap
is created
• Suspend – Maintain CR bitmap
• Resynch – Maintain PPRC mirror
• Planned HyperSwap
• Possibly for maintenance actions
Metro Mirror
(Synchronous PPRC)
PS
S
P
• Unplanned HyperSwap
• Swaps of complete disk configuration for various
problems that cause an interruption to the
communication between the server and the primary
disk subsystem
© Copyright IBM Corporation 2016
75
Remote Pair FlashCopy
• Function: Remote Pair FlashCopy (Preserve Mirror)
• Description: Remote Pair FlashCopy allows a Metro Mirror primary volume to
become the target of a FlashCopy relationship
• Why it matters: Without Remote Pair FlashCopy, a Metro Mirror primary
volume that is used as a target of a FlashCopy operation would cause all tracks
to have to be copied to the secondary. This would require more network
bandwidth for dataset level FlashCopy, volume copy, and system utilities that
utilize FlashCopy. In a HyperSwap environment, issuing a FlashCopy without
“preserve mirror” option would cause a “duplex pending” condition which would
temporarily disable HyperSwap until the process ended.
For more information, see IBM System Storage DS8000: Remote Pair FlashCopy, REDP-4504
© Copyright IBM Corporation 2016
76
Remote Pair FlashCopy
Without Remote Pair FlashCopy
• In this example, volume “Local B” is the target of a
FlashCopy operation. When the FlashCopy is
initiated, the Metro Mirror relationship is “duplex
pending” and HyperSwap is not permitted.
• All of the changed data on “Local B” must be sent over
the replication links to “Remote B”
© Copyright IBM Corporation 2016
With Remote Pair FlashCopy
• In the case where Remote Pair FlashCopy is used, the
FlashCopy is done from Local A to Local B and the
DS8000 will signal the secondary to perform the same
FlashCopy at the secondary site. The volumes are
always in full duplex mode and HyperSwap is
permitted
• No changed data due to the FlashCopy is sent across
the network, just the FlashCopy command to create a
FlashCopy at the secondary site
77
Soft Fence After Swap
• Function: Soft Fence After Swap
• Description: After a HyperSwap event, the old primary disk systems are
fenced to prevent a system from accidently accessing the now “offline” devices.
• Why it matters: After a planned or unplanned HyperSwap event, a potential
exposure exist to any systems connected to the old primary volumes. GDPS for
example uses this DS8000 capability to fence the devices to prohibit any host
system from inadvertently accessing or updating the wrong disk. This will
prevent any system wrongly IPL’ing from the wrong devices.
For more information, see IBM DS8870 and IBM z Systems Synergy, REDP-5186
© Copyright IBM Corporation 2016
78
Soft Fence After Swap
• After a swap event, the devices on the old
primary disk are fenced using a soft fence
to prohibit updates to the old control
unit(s).
• Soft fence protection will block any
attempt to update these volumes
© Copyright IBM Corporation 2016
79
Summary Event Notification (Metro Mirror Suspend)
• Function: Summary Event Notification (PPRC Suspend)
• Description: DS8870/DS8880 and current z/OS software levels support
message aggregation. When a replication session suspend occurs, the
DS8870/DS8880 aggregates state change interrupt messages.
• Why it matters: Without message aggregation, every PPRC device pair would
raise an alert that would be processed by z/OS due to state change
notifications. The DS8870/DS8880 provides message aggregation which
instead of each device raising a state change interrupt, the messages are
aggregated up to the logical control unit level, thereby reducing the message
traffic to a host to one message for 1 to 256 devices in a LCU.
For more information, see IBM DS8870 and IBM z Systems Synergy, REDP-5186
© Copyright IBM Corporation 2016
80
DS8870/DS8880 Health Message Alert
• Function: DS8870/DS8880 Health Message Alert
• Description: DS8870/DS8880 proactively monitors and reports when it is
experiencing an acute problem.
• Why it matters: Disk problems can be detected as a result of an I/O operation.
Additionally, the DS8870/DS8880 monitors and will report problems via the
health message alert capability and will alert z/OS whenever a hardware event
occurs. These notifications can be used as triggers by automation such as
GDPS and TPC-R to cause an unplanned HyperSwap event to protect
continued operations. The tight integration between the DS8870/DS8880 and
GDPS/TPC-R enables this capability.
For more information, see IBM DS8870 and IBM z Systems Synergy, REDP-5186
© Copyright IBM Corporation 2016
81
Data Protection and Backup
© Copyright IBM Corporation 2016
82
DFSMSdss Integration with FlashCopy
• Function: DFSMSdss Integration with FlashCopy
• Description: DFSMSdss, the z/OS data mover, can use FlashCopy functions if
requested
• Why it matters: Offloads both I/O and CPU processing from the z System to
the DS8000. Saves CPU cycles and I/O bandwidth
For more information, see IBM DS8870 and IBM z Systems Synergy, REDP-5186
© Copyright IBM Corporation 2016
83
DFSMSdss Integration with FlashCopy
• DFSMSdss Commands
BUILDSA
DEFRAG
CGCREATED
COMPRESS
CONSOLIDATE
CONVERTV
COPYDUMP
PRINT
RELEASE
COPY
DUMP
RESTORE
Items in BLUE exploit DS8000 FlashCopy (fast replication) functions
© Copyright IBM Corporation 2016
84
FlashCopy with DB2
• Function: FlashCopy with DB2
• Description: Standard DB2 utilities are integrated with DS8000 FlashCopy
• Why it matters: Utilizing FlashCopy with DB2 minimizes the application impact
when backup, restore and other utilities are running. Many DB2 utilities call
DFSMShsm (FRBACKUP/FRRECOV) which in turn uses FlashCopy services.
When using FlashCopy to move data, I/O processing is offloaded from the z
System to the disk system, saving processor CPU and I/O resources.
For more information, see DFSMShsm Fast Replication Technical Guide, SG24-7069
© Copyright IBM Corporation 2016
85
FlashCopy with DB2
DB2 Version
FlashCopy Exploitation
BACKUP SYSTEM
Version 8
RESTORE SYSTEM
Dataset FlashCopy support for CHECK INDEX SHRLEVEL CHANGE
Incremental FlashCopy for BACKUP SYSTEM
Version 9
Dataset FlashCopy for RECOVER with system-level backup (SLB) as input
Dataset FlashCopy for CHECK DATA SHRLEVEL CHANGE and CHECK LOB
SHRLEVEL CHANGE
Dataset FlashCopy for COPY
Dataset FlashCopy for inline copy in REORG TABLESPACE, REORG INDEX,
REBUILD INDEX, LOAD
Version 10
FlashCopy image copies with consistency and no application outage
(SHRLEVEL CHANGE)
FlashCopy Image Copy accepted as input to RECOVER, COPYTOCOPY,
DSN1COPY, DSN1COMP, DSN1PRNT
© Copyright IBM Corporation 2016
86
FlashCopy with IMS
• Function: FlashCopy with IMS
• Description: Standard IMS utilities are integrated with DS8000 FlashCopy
• Why it matters: IMS data management utilities, such as IMS High Performance
Image Copy, can use DS8000 FlashCopy functions, thus helping to shrink
maintenance windows and offloading I/O processing from the z System to the
DS8000.
For more information, see DFSMShsm Fast Replication Technical Guide, SG24-7069
© Copyright IBM Corporation 2016
87
FlashCopy with IMS
• IMS High Performance Image Copy
• Uses DFSMSdss DUMP and COPY commands
• DFSMSdss calls DS8000 FlashCopy
• Using FlashCopy minimizes application downtime to just the time that is required to
create a dataset level FlashCopy relationship then serialization is released
• Backup continues using FlashCopy copy of database
Both COPY and FDUMP use DS8000 FlashCopy function
DUMP uses DS8000 Concurrent Copy function
© Copyright IBM Corporation 2016
88
System z Continuous Data Protection for DB2 (zCDP)
• Function: System z Continuous Data Protection for DB2 (zCDP)
• Description: Application based continuous data protection for DB2 on System
z – a joint solution between DFSMS, DB2 and DS8000.
• Why it matters: A solution based on point-in-time backups combined with DB2
logging eliminates the need for DB2 log suspend. Can backup hundreds of
volumes in a matter of minutes and provides recovery at the system or
tablespace level.
For more information, see DFSMShsm Fast Replication Technical Guide, SG24-7069
© Copyright IBM Corporation 2016
89
System z Continuous Data Protection for DB2 (zCDP)
• DB2 system level backup and system level restore Backup calls HSM with DB Tables, HSM FlashCopy
to SMS Copy Pool, then DB Logs.
• DB2 Maintain Cross Volume Data Consistency. No quiesce of DB required.
• DFSMShsm function that manages Point-in-Time copies and integrated with DS8000 FlashCopy
• Combined with DB2 BACKUP SYSTEM, provides non-disruptive backup and recovery to any point in
time for DB2 and IMS databases and subsystems
CP Backup Storage Group
Copy Pool Application
Onsite
Offsite
FlashCopy
Dump to
Tape
Multiple
Disk Copies
Recovery at all levels from either disk or tape!
Entire copy pool, individual volumes and …
Individual data sets
© Copyright IBM Corporation 2016
• Up to 5 copies and 85 Versions of DB2 image
copies, managed by Management Class
•Automatic Expiration
90
T-10 Data Integrity Field
(aka T-10 Protection Information Model)
• Function: T-10 DIF for Linux on z Systems
• Description: This ANSI T10 standard provides a way to check the integrity of
data read and written from the host to the disk and back through the SAN fabric.
This check is implemented through the data integrity field (DIF) defined in the
T10 standard.
• Why it matters: Applications need to be absolutely sure that data stored on
disk when retrieved is exactly the data that was written. The SCSI Command
Descriptor Block enables this function by use of a data block guard, data block
application tag and a data block reference tag. The DS8000 as always, adds an
additional 4 bytes of existing meta data to ensure data integrity.
For more information, see IBM DS8880 Architecture and Implementation, SG24-8323
© Copyright IBM Corporation 2016
91
T-10 Data Integrity Field
(aka T-10 Protection Information Model)
• Provides advanced, end-toend data integrity for Linux
running on z Systems
• Checks the data integrity to
and from the host to the
disk through the SAN fabric
• Checking is done by
hardware – no performance
impact
© Copyright IBM Corporation 2016
92
IBM DS8000 and IBM z Systems Synergy
• Synergy is much more than just interoperability
• IBM owns the z System I/O architecture and z Systems and DS8870/DS8880 are
jointly developed
• IBM is best positioned for earliest delivery of new server support; sharing technology
between server and storage development
• No other disk vendor delivers 24/7 availability and optimized performance for z Systems
• Compatible ≠ identical.
• Other vendors support new z Systems features much later, or in some cases, never at
all
© Copyright IBM Corporation 2016
93
References
• DB2 for z/OS and List Prefetch Optimizer, REDP-4862
• http://www.redbooks.ibm.com/abstracts/redp4862.html?Open
• DFSMSdss Storage Administration, SC23-6868
• http://www-03.ibm.com/systems/z/os/zos/library/bkserv/v2r1pdf/
• DFSMShsm Fast Replication Technical Guide, SG24-7069
• https://www.redbooks.ibm.com/abstracts/sg247069.html?Open
• DS8000 I/O Priority Manager, REDP-4760
• http://www.redbooks.ibm.com/abstracts/redp4760.html?Open
• Get More Out of Your I/T Infrastructure with IBM z13 I/O Enhancements, REDP-5134
• http://www.redbooks.ibm.com/abstracts/redp5134.html?Open
• How Does the MIDAW Facility Improve the Performance of FICON, REDP-4201
• http://www.redbooks.ibm.com/abstracts/redp4201.html?Open
• IBM DS8880 Architecture and Implementation, SG24-8323
• http://www.redbooks.ibm.com/redpieces/abstracts/sg248323.html?Open
• IBM DS8870 Architecture and Implementation, SG24-8085
• http://www.redbooks.ibm.com/abstracts/SG248085.html?Open
• IBM DS8870 Copy Services for IBM z Systems, SG24-6787
• http://www.redbooks.ibm.com/abstracts/SG246787.html?Open
• IBM DS8870 and IBM z Systems Synergy, REDP-5186
• http://www.redbooks.ibm.com/abstracts/redp5186.html?Open
• IBM System Storage DS8000 Remote Pair FlashCopy (Preserve Mirror) REDP-4505
• http://www.redbooks.ibm.com/abstracts/redp4504.html?Open
• Effective zSeries Performance Monitoring Using Resource Measurement Facility, SG24-6645
• http://www.redbooks.ibm.com/abstracts/sg246645.html?Open
© Copyright IBM Corporation 2016
94
Additional Material
• IBM z13 and the DS8870 Series: Multi Target Metro Mirror and the IBM z13
https://www.youtube.com/watch?v=HokhHmAUhZY
• IBM z13 and the DS8870 Series: Fabric Priority
https://www.youtube.com/watch?v=o6cV7L14XSU
• IBM z13 and the DS8870 Series: zHyperWrite and DB2 Log Write Acceleration
https://www.youtube.com/watch?v=y96-cTwVHzs&index=3
• IBM z13 and the DS8870 Series: IBM FICON Dynamic Routing
https://www.youtube.com/watch?v=H70pZvR6EQo
• IBM z13 and the DS8870 Series: zHPF Extended Distance II
https://www.youtube.com/watch?v=pBEY-lYM2YY
© Copyright IBM Corporation 2016
95
Trademarks
The following are trademarks of the International Business Machines Corporation in the United States and/or other countries. For a complete list of IBM Trademarks, see
www.ibm.com/legal/copytrade.shtml: AS/400, DBE, e-business logo, ESCO, eServer, FICON, IBM, IBM Logo, iSeries, MVS, OS/390, pSeries, RS/6000, S/30, VM/ESA, VSE/ESA,
Websphere, xSeries, z/OS, zSeries, z/VM
The following are trademarks or registered trademarks of other companies
Lotus, Notes, and Domino are trademarks or registered trademarks of Lotus Development Corporation
Java and all Java-related trademarks and logos are trademarks of Sun Microsystems, Inc., in the United States and other countries
LINUX is a registered trademark of Linux Torvalds
UNIX is a registered trademark of The Open Group in the United States and other countries.
Microsoft, Windows and Windows NT are registered trademarks of Microsoft Corporation.
SET and Secure Electronic Transaction are trademarks owned by SET Secure Electronic Transaction LLC.
Intel is a registered trademark of Intel Corporation
* All other products may be trademarks or registered trademarks of their respective companies.
NOTES:
Performance is in Internal Throughput Rate (ITR) ratio based on measurements and projections using standard IBM benchmarks in a controlled environment. The actual
throughput that any user will experience will vary depending upon considerations such as the amount of multiprogramming in the user's job stream, the I/O configuration, the
storage configuration, and the workload processed. Therefore, no assurance can be given that an individual user will achieve throughput improvements equivalent to the
performance ratios stated here.
IBM hardware products are manufactured from new parts, or new and serviceable used parts. Regardless, our warranty terms apply.
All customer examples cited or described in this presentation are presented as illustrations of the manner in which some customers have used IBM products and the results
they may have achieved. Actual environmental costs and performance characteristics will vary depending on individual customer configurations and conditions.
This publication was produced in the United States. IBM may not offer the products, services or features discussed in this document in other countries, and the information
may be subject to change without notice. Consult your local IBM business contact for information on the product or services available in your area.
All statements regarding IBM's future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only.
Information about non-IBM products is obtained from the manufacturers of those products or their published announcements. IBM has not tested those products and cannot
confirm the performance, compatibility, or any other claims related to non-IBM products. Questions on the capabilities of non-IBM products should be addressed to the
suppliers of those products.
Prices subject to change without notice. Contact your IBM representative or Business Partner for the most current pricing in your geography.
References in this document to IBM products or services do not imply that IBM intends to make them available in every country.
Any proposed use of claims in this presentation outside of the United States must be reviewed by local IBM country counsel prior to such use.
The information could include technical inaccuracies or typographical errors. Changes are periodically made to the information herein; these changes will be incorporated in
new editions of the publication. IBM may make improvements and/or changes in the product(s) and/or the program(s) described in this publication at any time without notice.
Any references in this information to non-IBM Web sites are provided for convenience only and do not in any manner serve as an endorsement of those Web sites. The
materials at those Web sites are not part of the materials for this IBM product and use of those Web sites is at your own risk.
© Copyright IBM Corporation 2016
96
Disclaimers
Copyright © 2016 by International Business Machines Corporation.
No part of this document may be reproduced or transmitted in any form without written permission from IBM Corporation.
Product data has been reviewed for accuracy as of the date of initial publication. Product data is subject to change without notice. This
information could include technical inaccuracies or typographical errors. IBM may make improvements and/or changes in the
product(s) and/or programs(s) at any time without notice.
Any statements regarding IBM's future direction and intent are subject to change or withdrawal without notice, and represent goals and
objectives only.
References in this document to IBM products, programs, or services does not imply that IBM intends to make such such products,
programs or services available in all countries in which IBM operates or does business. Any reference to an IBM Program Product in
this document is not intended to state or imply that only that program product may be used. Any functionally equivalent program, that
does not infringe IBM’s intellectually property rights, may be used instead. It is the user’s responsibility to evaluate and verify the
operation of any on-IBM product, program or service.
THE INFORMATION PROVIDED IN THIS DOCUMENT IS DISTRIBUTED "AS IS" WITHOUT ANY WARRANTY, EITHER EXPRESS OR
IMPLIED. IBM EXPRESSLY DISCLAIMS ANY WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE OR
NONINFRINGEMENT. IBM shall have no responsibility to update this information. IBM products are warranted according to the terms
and conditions of the agreements (e.g., IBM Customer Agreement, Statement of Limited Warranty, International Program License
Agreement, etc.) under which they are provided. IBM is not responsible for the performance or interoperability of any non-IBM products
discussed herein.
The provision of the information contained herein is not intended to, and does not, grant any right or license under any IBM patents or
copyrights. Inquiries regarding patent or copyright licenses should be made, in writing, to:
IBM Director of Licensing
IBM Corporation
North Castle Drive
Armonk, NY 10504-1785
U.S.A.
© Copyright IBM Corporation 2016
97