Microsoft Word - White Paper - TS7700_R5.1Performance

IBM

TS7770, TS7770T, and TS7770C

Release 5.1

Performance White Paper

Version 2.0

By Khanh Ly

Virtual Tape Performance

IBM Tucson

IBM System Storage

™

December 24, 2020

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Table of Contents

Table of Contents ..............................................................................................................2

Introduction .......................................................................................................................3

Hardware Configuration ..................................................................................................4

TS7700 Performance Overview .......................................................................................5

TS7770 Basic Performance...............................................................................................8

TS7770 Grid Performance ............................................................................................. 11

Additional Performance Metrics ................................................................................... 22

Performance Tools .......................................................................................................... 33

Conclusions ...................................................................................................................... 34

Acknowledgements.......................................................................................................... 35

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Introduction

This paper provides performance information for the IBM TS7770, TS7770T, and

TS7770C, which are three current products in the TS7700 family of storage

products. This paper is intended for use by IBM field personnel and their

customers in designing virtual tape solutions for their applications.

This is an update to the previous TS7700 paper dated September 26, 2016 and

reflects changes for release 5.1, which features the TS7770, TS7770T, and

TS7770C. Some performance related features introduced since the last

performance white paper (TS7700 R 4.0 Performance White Paper) will also be

included.

Unless specified otherwise, in this white paper, all runs to a TS7770T target

a 300TB CP1 tape managed partition with ten FC 5274. All runs to a

TS7770C target a 300TB CP1 cloud managed partition with 4 FC5274.

The following are performance related changes since R 4.0:

 16Gb FICON support – R 4.1

 Software compression (ZSTD and LZ4) – R 4.1.2

 TS7700 cloud support - R 4.2

 Server Refresh to new Power9 pSeries - R 5.0

 Disk Cache Refresh to V5300 – R 5.0

 TS7700C Grid-awareness support – R 5.1

TS7770 Release 5.1 Performance White Paper Version 2 adds

data for the following configurations:

 Standalone TS7770T VED-T/CSB/n drawers (n=6, 8)

 TS7770T Premigration Rates vs Drawer Counts

 TS7770C Cloud Premigration Rate vs. Drawer Counts

 TS7770 Copy Performance Comparison

 Performance vs Block Size and Job counts

 Performance vs Compression Schemes and Job Counts

 Performance vs Compression Scheme and Block Sizes

 Virtual Mount Performance vs Configuration and Copy Modes

Product

Release

Enhancements

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Hardware Configuration

The following hardware was used in performance measurements. Performance

workloads are driven from IBM System zEC13 host with eight 16 Gb FICON

channels.

Standalone Hardware Setup

Grid Hardware Setup

IBM COS 3403

The following conventions are used in this paper:

TS7700 Drawer count Tape Lib/

Tape Drives

Cloud

IBM System

z™ Host

TS7770 VED

3956 CSB/XSB

2, 4, 6, 8, or 10

N/A N/A zEC13

TS7770T VED-T

3956 CSB/XSB

2, 4, or 10 TS4500/

12 TS1150

N/A

TS7770C VED-C

3956 CSB/XSB

2, 4, 6, 8, or 10

N/A

IBM COS

3403

TS7700

Drawer

count

Tape Lib/

Tape Drives

Cloud Grid links

(Gb)

IBM System

z™ Host

TS7770 VED

3956 CSB/XSB

10 N/A N/A

2x10 zEC13

TS7770T VED-T

3956 CSB/XSB

10 TS4500/

12 TS1150

N/A

2x10

TS7770T VED-C

3956 CSB/XSB

10 N/A IBM COS 3403

4x10

Maneger Accesser

Slicestor Gridlinks

/Accesser

Gridlinks/Slicestore

M01 2 A00 9 S01 2x10Gb 2x10Gb

Binary Decimal

Name

Symbol

Values in Bytes

Name

Symbol

Values in Bytes

kibibyte KiB 2

kilobyte KB 10

mebibyte

MiB

megabyte

gibibyte GiB 2

gigabyte GB 10

tebibyte

TiB

terabyte

pebibyte PiB 2

petabyte PB 10

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

TS7700 Performance Overview

Performance Workloads and Metrics

Performance shown in this paper has been derived from measurements that

generally attempt to simulate common user environments, namely a large number

of jobs writing and/or reading multiple tape volumes simultaneously. Unless

otherwise noted, all of the measurements were made with 128 simultaneously

active virtual tape jobs per active cluster. Each tape job was writing or reading

10.7 GiB of uncompressed data using 32 KiB blocks and QSAM BUFNO=20 that

compresses within the TS7770 at 5.351 using ZSTD compression. Measurements

were made with eight 16-gigabit (Gb) FICON channels on a zEC13 host. All runs

begin with the virtual tape subsystem inactive.

Unless otherwise stated, all runs were made with tuning values:

 DCOPYT=125,

 DCTAVGTD=100,

 ICOPYT=ENABLED,

 LINKSPEED=1000

 CPYPRIOR=DISABLED,

For TS7770T, there are additional settings:

 PMPRIOR=3600, PMTHLVL=4000,

 Reclaim disabled,

 Number of premigration drives per pool=10,

For TS7770C, there are additional settings:

 CPMCNTL=0

 CPMCNTH=60

 CDELDNT=16

 CLDPRIOR=3600 with only 4 FC5274 installed (notes: I only installed 4

FC5274 with CLDPRIOR=3600 so that I would have reasonable peak and

sustained periods in a 6-hour run).

Refer to the IBM® TS7700 Series Best Practices - Understanding, Monitoring

and Tuning the TS7700 Performance white paper for detailed description of

some of the different tuning settings.

Types of Throughput

The TS7770 or TS7770T

cp0

is a disk-cache only cluster, therefore read and write

data rates have been found to be fairly consistent throughout a given workload.

The TS7770T

cp1->7

contains physical tapes to which the cache data will be

periodically written and read, and therefore it exhibits four basic throughput rates:

peak write, sustained write, read-hit, and recall.

The TS7770C

cp1->7

connects to the cloud to which the cache data will be

periodically written and read, and therefore it also exhibits four basic throughput

rates: peak write, sustained write, read-hit, and recall.

Metrics and

Workloads

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Peak and Sustained Write Throughput.

For all TS7770T

cp1->7

measurements, any previous workloads have been allowed

to quiesce with respect to pre-migration to backend tape and replication to other

clusters in the grid. In other words, the test is started with the grid in an idle state.

Starting with this initial idle state, data from the host is first written into the

TS7770T

cp1->7

disk cache with little if any premigration activity taking place. This

allows for a higher initial data rate and is termed the “peak” data rate. Once a pre-

established threshold is reached of non-premigrated compressed data, the

amount of premigration is increased, which can reduce the host write data rate.

This threshold is called the premigration priority threshold (PMPRIOR), and has

default value of 1600 gigabytes (GB). When a second threshold of non-

premigrated compressed data is reached, the incoming host activity is actively

throttled to allow for increased premigration activity. This throttling mechanism

operates to achieve a balance between the amount of data coming in from the

host and the amount of data being copied to physical tape. The resulting data

rate for this mode of behavior is called the “sustained” data rate, and could

theoretically continue on forever, given a constant supply of logical and physical

scratch tapes. This second threshold is called the premigration throttling

threshold (PMTHLVL) and has a default value of 2000 gigabytes (GB). These two

thresholds can be used in conjunction with the peak data rate to project the

duration of the peak period. Note that both the priority and throttling thresholds

can be increased or decreased via a host command line request. For all the run

in this white paper, PMPRIOR and PMTHLVL were set to 3600 and 4000

respective to achieve a longer peak duration.

For all TS7770C

cp1->7

measurements, CLDPRIOR was set to 3600 to establish the

cloud premigration priority threshold. The premigration throttle threshold was

determined by the number the FC5274 installed. I only installed 4 FC5274 so that

there were reasonable peak and sustained periods in a 6-hour run).

Read-hit and Recall Throughput

Similar to write activity, there are two types of TS7770T

cp1->7

(or TS7770C

cp1->7

)

read performance: “read-hit” (also referred to as “peak”) and “recall” (also referred

to as “read-miss”). A read hit occurs when the data requested by the host is

currently in the local disk cache. A recall occurs when the data requested is no

longer in the disk cache and must be first read in from physical tape (or from the

cloud). Read-hit data rates are typically higher than recall data rates.

TS7770T

cp1->7

recall performance is dependent on several factors that can vary

greatly from installation to installation, such as number of physical tape drives,

spread of requested logical volumes over physical volumes, location of the logical

volumes on the physical volumes, length of the physical media, and the logical

volume size. Because these factors are hard to control in the laboratory

environment, recall is not part of lab measurement.

TS7770C

cp1->7

recall was not included in this white paper.

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Grid Considerations

Up to five TS7700 clusters can be linked together to form a grid configuration.

Six-, seven-, and eight-way grid configurations are available via iRPQ. The

connection between these clusters is provided by two 1-Gb, four 1-Gb links, two

10-Gb links, or four 10-Gb TCP/IP links. Data written to one TS7700 cluster can

be optionally copied to the one or more other clusters in the grid.

Data can be copied between the clusters in either deferred, RUN (also known as

“Immediate”), or sync mode copy. When using the RUN copy mode the rewind-

unload response at job end is held up until the received data is copied to all peer

clusters with a RUN copy consistency point. In deferred copy mode data is

queued for copying, but the copy does not have to occur prior to job end if DCT is

set to zero (default). Deferred copy mode allows for a temporarily higher host

data rate than RUN copy mode because copies to the peer cluster(s) can be

delayed, which can be useful for meeting peak workload demands. Care must be

taken, however, to be certain that there is sufficient recovery time for deferred

copy mode so that the deferred copies can be completed prior to the next peak

demand. Whether delay occurs and by how much is configurable through the

Library Request command. In sync mode copy, data synchronization is up to

implicit or explicit sync point granularity across two clusters within a grid

configuration. In order to provide a redundant copy of these items with a zero

recovery point objectives (RPO), the sync mode copy function will duplex the host

record writes to two clusters simultaneously.

Grid

Considerations

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

TS7770 Basic Performance

The following sets of graphs show basic TS7770 bandwidths. The graphs in

Figures 1, 2, and 3 show single cluster, standalone configurations. Unless

otherwise stated, the performance metric shown in these and all other data rate

charts in this paper is host-view (uncompressed) MB/sec.

TS7770 Standalone Performance

Figure 1. TS7770 Standalone Maximum Host Throughput. All runs were made

with 128 concurrent jobs, each job writing and/or reading 2000 MiB (with 1:1

compression) or 10.7 GiB (with 5.35:1 compression), using 32KiB blocks, QSAM

BUFNO = 20, using eight 16Gb (8x16Gb) FICON channels from a zEC13 LPAR.

Notes:

Mixed 1:1 workload refers to a host pattern made up of 50% jobs which

read and 50% jobs which write. The resulting read and write activity

measured in the TS7770 varied and was rarely exactly 50/50

Mixed 5:35:1 workload refers to host pattern made up of 25% read jobs

and 75% write jobs.

TS7770

Standalone

Maximum Host

Throughput

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

TS7770T

cp1

Standalone Performance

Figure 2. TS7770T

cp1

Standalone Maximum Host Throughput. All runs were

made with 128 concurrent jobs, each job writing and/or reading 2000 MiB (with

1:1 compression) or 10.7 GiB (with 5.35:1 compression), using 32KiB blocks,

QSAM BUFNO = 20, using eight 16Gb (8x16Gb) FICON channels from a zEC13

LPAR.

Notes:

Mixed 1:1 workload refers to a host pattern made up of 50% jobs which

read and 50% jobs which write. The resulting read and write activity

measured in the TS7770T varied and was rarely exactly 50/50

Mixed 5:35:1 workload refers to host pattern made up of 25% read jobs

and 75% write jobs.

TS7770T

cp1->7

Standalone

Maximum Host

Through

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

TS7770C

cp1

Standalone Performance

Figure 3. TS7770C

cp1

Standalone Maximum Host Throughput. All runs were

made with 128 concurrent jobs, each job writing and/or reading 2000 MiB (with

1:1 compression) or 10.7 GiB (with 5.35:1 compression), using 32KiB blocks,

QSAM BUFNO = 20, using eight 16Gb (8x16Gb) FICON channels from a zEC13

LPAR.

Notes:

Mixed 1:1 workload refers to a host pattern made up of 50% jobs which

read and 50% jobs which write. The resulting read and write activity

measured in the TS7770C varied and was rarely exactly 50/50

Mixed 5:35:1 workload refers to host pattern made up of 25% read jobs

and 75% write jobs.

TS7770C

cp1->7

Standalone

Maximum Host

Throughput

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

TS7770 Grid Performance

Figures 4 through 6, 8 through 11, 13, 15, 17, and 19 display the performance for

TS7770 grid configurations.

For these charts “D” stands for deferred copy mode, “S” stands for sync mode

copy and “R” stands for RUN (immediate) copy mode. For example, in Figure 4,

RR represents RUN for cluster 0, and RUN for cluster 1. SS refers to

synchronous copies for both clusters.

All measurements for these graphs were made at zero or near-zero distance

between clusters.

Two-way TS7770 Grid with Single Active Cluster Performance

Figure 4. Two-way TS7770 Single Active Maximum Host Throughput. Unless

otherwise stated, all runs were made with 128 concurrent jobs. Each job writing

10.7 GiB (2000 MiB volumes @ 5.35:1 compression) using 32 KiB block size,

QSAM BUFFNO = 20, using eight 16Gb FICON channels from a zEC13 LPAR.

Clusters are located at zero or near zero distance to each other in laboratory

setup. DCT=125.

Notes:

SDT/AES-256 (Security Data Transfer with TLS 1.2 AES256): Encrypted user

data for grid replication.

Two-way TS7770

Grid Single Active

Maximum Host

Throughput

TS7700 Grid

Maximum Host

Throughput

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Figure 5. Two-way TS7770T

cp1

Single Active Maximum Host Throughput.

Unless otherwise stated, all runs were made with 128 concurrent jobs. Each job

writing 10.7 GiB (2000 MiB volumes @ 5.35:1 compression) using 32 KiB block

size, QSAM BUFFNO = 20, using eight 16Gb FICON channels from a zEC13

LPAR. Clusters are located at zero or near zero distance to each other in

laboratory setup. DCT=125.

Notes:

SDT/AES-256 (Security Data Transfer with TLS 1.2 AES256): Encrypted user

data for grid replication.

Two-way

TS7770T

cp1

Grid

Single Active

Maximum Host

Throughpu

Two

way

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Figure 6. Two-way TS7770C

cp1

Single Active Maximum Host Throughput.

Unless otherwise stated, all runs were made with 128 concurrent jobs. Each job

writing 10.7 GiB (2000 MiB volumes @ 5.35:1 compression) using 32 KiB block

size, QSAM BUFFNO = 20, using eight 16Gb FICON channels from a zEC13

LPAR. Clusters are located at zero or near zero distance to each other in

laboratory setup. DCT=125.

Notes:

* HTTPs: Communication protocol between TS7770C and cloud.

* SDT/AES-256 (Security Data Transfer with TLS 1.2 AES256): Encrypted user

data for grid replication.

Two-way

TS7770C

cp1

Grid

Single Active

Maximum Host

Throughpu

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Figure 7. Two-way TS7700 Hybrid Grid H1

Figure 8. Two-way TS7770-TS7770T

cp1

Hybrid H1 Single Active Maximum Host

Throughput. Unless otherwise stated, all runs were made with 128 concurrent

jobs. Each job writing 10.7 GiB (2000 MiB volumes @ 5.35:1 compression)

using 32 KiB block size, QSAM BUFFNO = 20, using eight 16Gb FICON

channels from a zEC13 LPAR. Clusters are located at zero or near zero

distance to each other in laboratory setup. DCT=125.

Two-way TS7770

Hybrid Grid H1

Single Active

Maximum Host

Throughpu

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Two-way TS7700 Grid with Dual Active Clusters Performance

Figure 9. Two-way TS7770 Dual Active Maximum Host Throughput. Unless

otherwise stated, all runs were made with 256 concurrent jobs (128 jobs per

active cluster). Each job writing 10.7 GiB (2000 MiB volumes @ 5.35:1

compression) using 32 KiB block size, QSAM BUFFNO = 20, using eight 16Gb

FICON channels from a zEC13 LPAR. Clusters are located at zero or near zero

distance to each other in laboratory setup. DCT=125.

Notes:

SDT/AES-256 (Security Data Transfer with TLS 1.2 AES256): Encrypted user

data for grid replication.

Two-way TS7700

Grid Dual Active

Maximum Host

Throughput

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Figure 10. Two-way TS7770T

cp1

Dual Active Maximum Host Throughput.

Unless otherwise stated, all runs were made with 256 concurrent jobs (128 jobs

per active cluster). Each job writing 10.7 GiB (2000 MiB volumes @ 5.35:1

compression) using 32 KiB block size, QSAM BUFFNO = 20, using eight 16Gb

FICON channels from a zEC13 LPAR. Clusters are located at zero or near zero

distance to each other in laboratory setup. DCT=125.

Notes:

SDT/AES-256 (Security Data Transfer with TLS 1.2 AES256): Encrypted user

data for grid replication.

Two-way

TS7700T

cp1

Grid

Dual Active

Maximum Host

Throughput

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Figure 11. Two-way TS7770C

cp1

Dual Active Maximum Host Throughput.

Unless otherwise stated, all runs were made with 256 concurrent jobs (128 jobs

per active cluster). Each job writing 10.7 GiB (2000 MiB volumes @ 5.35:1

compression) using 32 KiB block size, QSAM BUFFNO = 20, using eight 16Gb

FICON channels from a zEC13 LPAR. Clusters are located at zero or near zero

distance to each other in laboratory setup. DCT=125.

Notes:

* HTTPs: Communication protocol between TS7770C and cloud.

* SDT/AES-256 (Security Data Transfer with TLS 1.2 AES256): Encrypted user

data for grid replication.

Two-way

TS7700C

cp1

Grid

Dual Active

Maximum Host

Throughput

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Figure 12. Two-way TS7700 Hybrid Grid H2

Figure 13. Two-way TS7770-TS7770T

cp1

Hybrid H2 Dual Active Maximum Host

Throughput. Unless otherwise stated, all runs were made with 256 concurrent

jobs (128 jobs per active cluster). Each job writing 10.7 GiB (2000 MiB volumes

@ 5.35:1 compression) using 32 KiB block size, QSAM BUFFNO = 20, using

eight 16Gb FICON channels from a zEC13 LPAR. Clusters are located at zero

or near zero distance to each other in laboratory setup. DCT=125.

Two-way TS7770

Hybrid Grid H2

Dual Active

Maximum Host

Throughpu

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Figure 14. Three-way TS7700 Hybrid Grid H4

Figure 15. Three-way TS7770-TS7770T

cp1

Hybrid H4 Dual Active Maximum

Host Throughput. Unless otherwise stated, all runs were made with 256

concurrent jobs (128 jobs per active cluster). Each job writing 10.7 GiB (2000

MiB volumes @ 5.35:1 compression) using 32 KiB block size, QSAM BUFFNO =

20, using eight 16Gb FICON channels from a zEC13 LPAR. Clusters are located

at zero or near zero distance to each other in laboratory setup. DCT=125.

Three-way TS7770

Hybrid Grid H4

Dual Active

Maximum Host

Throughpu

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Figure 16. Four-way TS7700 Hybrid Grid H6

Figure 17. Four-way TS7770-TS7770T

cp1

Hybrid H6 Dual Active Maximum Host

Throughput. Unless otherwise stated, all runs were made with 256 concurrent

jobs (128 jobs per active cluster). Each job writing 10.7 GiB (2000 MiB volumes

@ 5.35:1 compression) using 32 KiB block size, QSAM BUFFNO = 20, using

eight 16Gb FICON channels from a zEC13 LPAR. Clusters are located at zero

or near zero distance to each other in laboratory setup. DCT=125.

Four-way TS7770

Hybrid Grid H6

Dual Active

Maximum Host

Throughpu

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Figure 18. Four-way TS7700 Hybrid Grid H7

Figure 19. Four-way TS7770-TS7770T

cp1

Hybrid H7 Four Active Maximum Host

Throughput. Unless otherwise stated, all runs were made with 256 concurrent

jobs (128 jobs per active cluster). Each job writing 10.7 GiB (2000 MiB volumes

@ 5.35:1 compression) using 32 KiB block size, QSAM BUFFNO = 20, using

eight 16Gb FICON channels from a zEC13 LPAR. Clusters are located at zero

or near zero distance to each other in laboratory setup. DCT=125.

Four-way TS7700

Hybrid H7

Quadruple Active

Maximum Host

Throughput

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Additional Performance Metrics

TS7760 Performance vs. FICON Channel Configuration

The figure 20 shows how the number and/or configuration of the FICON

channels affects host throughput.

Figure 20. TS7770 Standalone Maximum Host Throughput. All runs were made

with 128 concurrent jobs, each job writing 10.7 GiB (with 5.35:1 compression),

using 32KiB blocks, QSAM BUFNO = 20, using eight 16Gb (8x16Gb) FICON

channels from a zEC13 LPAR.

Host Data Rate vs.

FICON Channel

Configuration

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

TS7770T Sustained and Premigration Rates vs. Premigration

Drives

TS7770T

cp1->7

premigration rates, i.e. the rates at which cache-resident data is

copied to physical tapes, depend on the number of TS1150 tape drives

reserved for premigration and the number of disk drawers installed. By

default, the number of tape drives reserved for premigration is ten per pool.

TS7770T

cp1->7

sustained write rate is the rate at which host write rate balanced

with premigration to tape, also depends on the number of premigration tape

drives.

The figure 21 shows how the number of premigration tape drives affects

premigration rate and sustained write rate.

Figure 21. Standalone TS7770T

cp1

sustained write rate and tape premigration

rate vs. the number of TS1150 drives reserved for premigration. All runs were

made with 128 concurrent jobs. each job writing 10.7 GiB (with 5.35:1

compression), using 32KiB blocks, QSAM BUFNO = 20, using eight 16Gb

(8x16Gb) FICON channels from a zEC13 LPAR.

Sustained Host

rate and

Premigration rates

vs. Premigration

Tape Drives

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

TS7770T Premigration Rates vs. Drawer Counts

The figure 22 shows that the number of cache drawers affects premigration

rate (with and without host activity).

Figure 22. Standalone TS7770T

cp1

sustained write rate and tape premigration

rate vs. the number of cache drawers. All runs were made with 128 concurrent

jobs. each job writing 10.7 GiB (with 5.35:1 compression), using 32KiB blocks,

QSAM BUFNO = 20, using eight 16Gb (8x16Gb) FICON channels from a zEC13

LPAR.

Sustained Host

rate and

Premigration rates

vs. Cache Drawer

Counts

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

TS7770C Cloud Premigration Rates vs. Drawer Counts

The figure 23 shows that the number of cache drawers affects cloud

premigration rate (with and without host activity).

Figure 23. Standalone TS7770C

cp1

sustained write rate and cloud premigration

rate vs. the number of cache drawers. All runs were made with 128 concurrent

jobs. each job writing 10.7 GiB (with 5.35:1 compression), using 32KiB blocks,

QSAM BUFNO = 20, using eight 16Gb (8x16Gb) FICON channels from a zEC13

LPAR.

Sustained Host

Rate and Cloud

Premigration

Rates vs. Cache

Drawer Counts

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

TS7700 Copy Performance Comparison

In each of the following runs, a deferred copy mode run was ended following

several terabyte (TB) of data being written to the active cluster(s). In the

subsequent hours, copies took place from the source cluster to the target cluster.

There was no other TS7700 activity during the deferred copy.

Figure 24. Two-way TS7700 Maximum Copy Throughput. Clusters are located

at zero or near zero distance to each other in laboratory setup.

Notes:

SDT/AES-256 (Security Data Transfer with TLS 1.2 AES256): Encrypted user

data for grid replication.

TS7700 Copy

Performance

Comparison

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

For 2-way TS7700T

cp1

, the premigration activity on the source/target cluster

consumes resources and thus lower the copy performance on the TS7700T

as compared to the TS7700.

Figure 25. Two-way TS7700T

cp1

Maximum Copy Throughput. Clusters are

located at zero or near zero distance to each other in laboratory setup.

Notes:

SDT/AES-256 (Security Data Transfer with TLS 1.2 AES256): Encrypted user

data for grid replication.

TS7700T

Sustained Copy

Performance

Evolution

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Performance vs. Block size and Number of Concurrent Jobs.

Figure 26 shows data rates on a standalone TS7770 VED/CSB/10

drawers/8x16Gb FICON with different job counts driven from a zEC13 host

using different channel block sizes. Significant performance improvement

occurs using 64KB and 128KB block size with 64 concurrent jobs.

Figure 26. TS7770 Standalone Maximum Host Throughput. All runs were made

with 128 concurrent jobs. each job writing 10.7 GiB (with 5.35:1 compression),

with QSAM BUFNO = 20, using eight 16Gb (8x16Gb) FICON channels from a

zEC13 LPAR.

Standalone

TS7770

Performance vs

Block Sizes and

Job Counts

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Performance vs. Compression Schemes and Job Counts

Figures 27 and 28 show data rates on a standalone TS7770 VED/CSB/10

drawers/8x16Gb FICON with different workload job counts driven from a

zEC13 host using different TS7770 compression schemes (FICON, LZ4, and

ZSTD). The difference between figure 27 and figure 28 was the block size

used (32KB vs 256KB).

Figure 27. TS7770 Standalone Maximum Host Throughput. All runs were made

with 128 concurrent jobs. each job writing 10.7 GiB (with 5.35:1 compression),

with 32KB block size and QSAM BUFNO = 20, using eight 16Gb (8x16Gb)

FICON channels from a zEC13 LPAR.

Standalone TS7770

Performance vs

Compression

Schemes and Job

Counts

(32KB Block Size)

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Figure 28. TS7770 Standalone Maximum Host Throughput. All runs were made

with 128 concurrent jobs. each job writing 10.7 GiB (with 5.35:1 compression),

with 256KB block size and QSAM BUFNO = 20, using eight 16Gb (8x16Gb)

FICON channels from a zEC13 LPAR.

Standalone TS7770

Performance vs

Compression

Schemes and Job

Counts

(256KB Block Size)

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Performance vs. Compression Schemes and Block Sizes

Figure 29 shows how data compression ratio varies depending on the

compression scheme and block size used.

Figure 29. TS7770 Compression Ratio. All runs were made with 128 concurrent

jobs. each job writing 10.7 GiB (with 5.35:1 compression), with QSAM BUFNO =

20, using eight 16Gb (8x16Gb) FICON channels from a zEC13 LPAR.

Standalone TS7770

Performance vs

Compression

Schemes and Block

Sizes

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Virtual Mount Performance vs. Configurations and Copy Modes

In each of the following mount intensive runs, very small volumes were used to

obtain the highest number of virtual mounts.

Figure 30. TS7770 Maximum Mount Rate. All runs were made with 128

concurrent jobs. each job writing 100 MiB (with 5.35:1 compression), with 32KB

block size and QSAM BUFNO = 20, using eight 16Gb (8x16Gb) FICON channels

from a zEC13 LPAR.

TS7700 Maximum

Virtual Mount vs

Hardware

Configurations and

Copy Modes

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Performance Tools

Batch Magic

This tool is available to IBM representatives and Business Partners to analyze

SMF data for an existing configuration and workload, and project a suitable

TS7700 configuration.

BVIRHIST plus VEHSTATS

BVIRHIST requests historical statistics from a TS7700, and VEHSTATS

produces the reports. The TS7700 keeps the last 90 days of statistics.

BVIRHIST allows users to save statistics for periods longer than 90 days.

Performance Analysis Tools

A set of performance analysis tools is available on Techdocs that utilizes the data

generated by VEHSTAT. Provided are spreadsheets, data collection

requirements, and a 90 day trending evaluation guide to assist in the evaluation

of the TS7700 performance. Spreadsheets for a 90 day, one week, and a 24

hour evaluation are provided.

http://www-03.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/PRS4717

Also, on the Techdocs site is a webinar replay that teaches you how to use the

performance analysis tools.

http://www-03.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/PRS4872

BVIRPIT plus VEPSTATS

BVIRPIT requests point-in-time statistics from a TS7700, and VEPSTATS

produces the reports. Point-in-time statistics cover the last 15 seconds of activity

and give a snapshot of the current status of drives and volumes.

The above tools are available at one of the following web sites:

ftp://public.dhe.ibm.com/storage/tapetool/

Performance Aids

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Conclusions

The TS7700 provides significant performance improvement, increased capacity,

and new functionality over the years. Release 4.1 introduced 16Gb FICON

channel support which increased the maximum channel performance from 2500

MB/s to over 4000 MB/s. Release R 4.1.2 introduced software compression LZ4

and ZSTD which increase the compression ratio very significantly as compared

to the traditional hardware compression at the FICON adapter level (FICON

compression). Release 4.2 introduced cloud support. Release 5.0 introduces

new TS7770, TS7770T, and TS7770C models with new Power 9 server and

V5000 cache. The TS7700 architecture provides a base for product growth in

both performance and functionality.

Conclusions

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

Acknowledgements

The author would like to thank Joseph Swingler, and Donald Denning for their

review comments and insight, and also to Gary Anna for publishing the paper.

The author would like to thank Albert Veerland for Performance Driver support.

Finally, the author would like to thank Donald Denning, James F. Tucker,

Kymberly Beeston, Dennis Martinez, George Venech, and Jeffrey Watson for

hardware/zOS/network support.

Acknowledgements

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

IBM Systems

9000 South Rita Road

Tuson, AZ 85744

Printed in the United States of America

12-20

IBM, the IBM logo, System Storage, System z, zOS, TotalStorage, ,

DFSMSdss, DFSMShsm, ESCON and FICON are trademarks or

registered trademarks of International Business Machines

Corporation in the United States, other countries, or both.

Other company, product and service names may be trademarks or

service marks of others.

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.

Performance data for IBM and non-IBM products and services

contained in this document were derived under specific operating

and environmental conditions. The actual results obtained by any

party implementing such products or services will depend on a large

number of factors specific to such party’s operating environment

and may vary significantly. IBM makes no representation that these

results can be expected or obtained in any implementation of any

such products or services.

References in this document to IBM products, programs, or services

does not imply that IBM intends to make such products, programs

or services available in all countries in which IBM operates or does

business.

IBM TS7700 Release 5.1 Performance White Paper Version 2.0

Page

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 INFRINGEMENT.

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.