For many larger enterprises, the pressure to squeeze more performance out of existing mainframes, while avoiding or deferring major upgrades, is a top priority. Routine sort processing tasks such as sort, copy, merge, and compression take up a large and disproportionate share of CPU processing time and related mainframe resources. They are thus obvious areas to examine when seeking relief from mainframe cost and performance pressures.
When evaluating potential solutions many organizations want to “try before they buy” to ensure strong ROI. We have developed tools to help you understand your batch processing and calculate your expected ROI.
Watch this on-demand webinar to hear about:
• How our Syncsort MFX solution drive cost savings
• What a “try before you buy” evaluation looks like
• How to analyze results for your specific situation
Unlock Cost Savings for your IBM z Systems Environment
1. Unlock cost savings
for your IBM z system
environment
Bill Hammond | Director, Product Marketing
Alissa Margulies | Senior Sales Engineer
2. Housekeeping
Webinar Audio
• Today’s webcast audio is streamed through your
computer speakers
• If you need technical assistance with the web interface
or audio, please reach out to us using the Q&A box
Questions Welcome
• Submit your questions at any time during the
presentation using the Q&A box. If we don't get to your
question, we will follow-up via email
Recording and slides
• This webinar is being recorded. You will receive an email
following the webinar with a link to the recording and
slides
3. Today’s topics
• Mainframe customer priorities
• Precisely can help
• Predicting your benefits
• Customer stories
• Q & A
3
5. Mainframes host the most critical applications
71% 2.5 Billion $2.9 Billion
92 10 23
5
6. * Arcati Mainframe Yearbook 2023
** Zetaly Mainframe Cost Control Blog post
6
Mainframes expenses
represent between
20 and 40% of
companies’ IT budget**
70% of sites think the biggest
obstacle for mainframe
acceptance is perception
of being too expensive*
7. Optimizing SORT
• Sort operations (SORT, MERGE, COPY) are very
resource intensive:
• Significant CPU cycles needed to sort records
using sorting algorithms
• I/O intensive – read input, read/write SORTWK,
write output data sets
Focus on reducing CPU time and
I/O efficiency for sort operations
7
8. IBM zIIP engines
IBM® z Integrated Information Processor (zIIP):
• A purpose-built processor designed to operate asynchronously
with the general processors in the mainframe to help improve
utilization of computing capacity and control costs.
• In use at most mainframe customers … up to 85% adoption*
• zIIPs allow customers to purchase additional processing power
without affecting the total million service units (MSU) rating
• IBM does not impose IBM software charges on zIIP capacity,
but charges apply when additional general purpose engines capacity
is used
Bottom line:
• Once zIIPs are purchased they are used without any additional cost
• Moving eligible workloads to zIIP reduces general purpose engine
utilization and can save SW license costs
• The big challenge is that most workloads are not enabled to run
on zIIP
* Arcati Mainframe Yearbook 2023
8
10. Syncsort MFX
Over 50 years of continual
development and enhancements
Performance
The high-performance
IBM zSystems sort solution
that delivers better
performance and saves
you money
Proven Solution
Improves sort performance while
optimizing overall system efficiency
zIIP Offload
Sort workloads can be directed to
the zIIP, thereby lowering the CPU
time and costs
Encryption
Enhanced security and compliance
with regulations such as GDPR
10
11. Reduced CPU time
• Sort operations are resource intensive
• Use less CPU time
• Optimize job run times
11
12. Control MLC costs
• Significantly reduces general processor
operations
• Allows sort work encryption to be
completed on zIIP
• Dramatically reduces CPU utilization
and costs
12
13. Delay hardware
upgrades
• Delays hardware upgrades even as
data volumes and workloads increase
• Eliminates need for Capacity on
Demand expenses
13
15. Justification for
zIIP expenses
• Offloads a significant amount of sorting
workload to zIIP engines
• Makes use of under-utilized zIIP engines
15
16. A powerful combination
Syncsort MFX focuses on
reducing CPU time and I/O
activity for sort operations
Syncsort ZPSaver zIIP-enables sort
operations to control MLC costs
and delay/prevent upgrades
• Significantly reduces general
processor operation
• Makes use of under-utilized
zIIP engines
• Helps control MLC costs and
delay/prevent CPU upgrades
• Allows sort work encryption to
be completed on zIIP, reducing
CPU utilization dramatically and
reducing costs
• Sort operations are resource
intensive
• Syncsort MFX uses less CPU time
than DFSORT
• Syncsort MFX optimizes I/O better
than DFSORT
• Syncsort MFX encrypts sort work
for required security
and compliance
• Faster execution means more work
can execute in the same amount
of time
Syncsort MFX PipeSort reduces
elapsed time for sort operations
• Simultaneously executes up
to eight differently sequenced
sorts from a single pass of the
input data
• Uses advanced parallel sorting
technology
• Cuts total elapsed time by more
than 50% compared
to running separate sorts
16
18. Syncsort MFX key features
• Dynamic Storage Management (DSM) monitors and
dynamically controls sort performance and resource use
• Dynamic sort work allocation technique acquires sort work as
required during the sort step instead of allocating all anticipated
space at sort initiation
• Data Editing features perform a range of data manipulation
functions without additional programming
• Encrypts sort work data sets to provide enhanced security and
compliance with regulations such as GDPR
• Available Features
• Syncsort ZPSaver - Offload 90% of your Copy, SMS Compression,
and Sort CPU cycles to zIIP
• Syncsort PipeSort - offers the fastest way of sorting the same file
many times with different key sequences
• Syncsort PROCSort - a high-performance, transparent replacement
for the SAS-provided sort procedure PROCSort
18
23. Leading
telecommunication
firm
“Syncsort MFX allowed us to
avoid a costly upgrade in
capacity on demand and
also lower software costs.”
— Senior technical analyst
O b j e c t i v e
• Complete required billing work during the
daily batch window without increasing
mainframe hardware and software
licensing costs
C h a l l e n g e
• 100% CPU during batch window
• Jobs impacted and work not completing
during batch window
• How to fix the situation without costly
upgrades
S o l u t i o n
• Syncsort MFX + Syncsort ZPSaver, the
zIIP offload solution
B e n e f i t
• Immediate reduction in CPU time and
elapsed time
• Reduction in CPU time in batch
window prevented increasing capacity
on demand
• Jobs completed in batch windows,
avoiding impacting other work
• Reduction in software cost due to a
decrease in consumed MSUs
23
24. International
credit card
processor
Found $1,500,000 in annual
savings by deploying
Syncsort ZPSaver
O b j e c t i v e
• Drive down mainframe costs via
efficiencies such as zIIP offload,
tuning, elimination of work or
offload to distributed systems
C h a l l e n g e
• Cost of any software solution must be
offset by hard dollar savings
• Cost savings can be achieved by
substantial reduction of MSUs in peak
4HRA window which occurs during
batch processing
• As CPU is freed in peak window latent
demand moves in and negates
savings
S o l u t i o n
• Syncsort ZPSaver showed average
savings of 35 MSUs per hour in peak
4HRA window translating to an annual
savings of $1,500,000 in MLC
• A 2nd POC was underway to address
the concern with latent demand by
gradually lowering the defined MSU cap
• The lengthy 2nd POC was cut short with
offer of a special bundled price with the
MFX renewal that lowered the
justification requirements
B e n e f i t
• Reduction in mainframe MLC costs via
zIIP offload during peak 4HRA window
• Transparent to existing sort jobs; no
changes needed
• An easier justification and acquisition
process within procurement cycle for the
renewal
24
25. Major
automobile
manufacturer
Efficiency gained with
Syncsort MFX & Syncsort
ZPSaver resulted in delaying
a hardware upgrade for
almost and entire year
O b j e c t i v e
• Extend trusted relationship with
Precisely by exploring another
solution, in addition to Syncsort
Optimize IMS, to address cost
reduction goals for mainframe
environment
C h a l l e n g e
• Was tasked with reducing
mainframe costs but had no spare
manpower resources to dedicate to
tuning projects
• Growing workload and data
volumes in batch processes were
driving need for hardware upgrades
• Under-utilized zIIP processors
required post-acquisition
justification of the hardware
investment
S o l u t i o n
• Syncsort MFX + Syncsort ZPSaver
transparently replaced IBM
DFSORT as the system sort utility
with little investment of systems
programming time and effort
• Reductions in both CPU and
elapsed times in sort-intensive
batch workload allowed additional
work and larger files to be
processed with existing capacity
B e n e f i t
• Capacity projections pushed
hardware upgrade out by almost a
year delaying both hardware and
software upgrade expenses
• Significant zIIP offload more than
justified financial investment in
specialty engines
25
26. Sort performance and
tuning service
We can help you…
• Understand the factors that affect sort and system
performance
• Review and tune existing applications to meet your
performance objectives
• Learn how to keep your configuration tuned as their
environment evolves
26
29. Syncsort MFX
• Offers a high-performance sort, copy, and join
utility designed to exploit the advanced facilities
of the z/OS operating system and IBM Z, including
zIIP engines.
• Can significantly reduce CPU utilization for sort
operations and optimize I/O activity to reduce
contention to control software costs and delay
CPU upgrades
• Frees up general purpose MIPS for handling
increased data volumes and new workloads
• Drive cost reduction strategies by delaying CPU
upgrades and reducing software charges.
29
Notas del editor
Mainframes are still the backbone for the biggest organizations in the world
71% of the fortune 500 rely on the mainframe for their mission critical transactional systems and they are present in every vertical from FinServ to Insurance to Retail.
When talking to these organizations, it’s not unusual to hear that up to 80% of their corporate data originates on the mainframe and that business is growing. The mainframe market is expected to grow to $2.9 billion by 2025.
CPU cycles used
Potentially raises LPAR MSU total consumption
Could result in increased pricing
Takes CPU cycles from other critical workloads
I/O intensive – read input, read/write SORTWK, write output data sets
Typically take long to execute because storage devices are not as fast as processors
I/O to a device is typically single-threaded
Contention and bottlenecks can result when workloads compete for the same storage devices
Long running workloads may impact dependent workloads down-stream
MFX focuses on reducing CPU time and I/O activity for sort operations!
Challenge:
With CPU utilization peaking at 100% in the daily batch window when critical billing jobs were running, this telecommunications firm was seeing the performance of the jobs being impacted, along with an occasional inability to complete all jobs within the window. They were looking at an impending upgrade which they simply didn’t want to budget for or take on. Furthermore, their 4-Hour Rolling Average (4HRA) MSU peak utilization was creeping upward driving MLC costs up with it and they had to find a solution to reduce software costs by reducing MSUs in that peak period.
Solution:
The company took advantage of a free Precisely SMF analysis to get a projection on how much they could save in terms of MIPS (and ultimately MSUs and MLC costs), and in elapsed time within their batch window, with Precisely’s industry-leading sort and zIIP offload solution ZPSaver. A flexible analysis, tailored to their specific needs, was performed to provide them with a clear picture of where ZPSaver could produce savings in general CP utilization, as well as reduce elapsed time, providing a justification for purchase. One analysis was done for all systems during all times of day, in another a refined analysis focused on the 4HRA peak window, and finally an analysis of several days where 100% CPU utilization was hit during the batch window was performed.
Results
This comprehensive analysis projected results that were astounding with an average savings of 72.4% in CPU time along with an 11% savings in elapsed time within their current batch window. After a quick install the company saw an immediate reduction in CPU time and elapsed time. Most critical was the reduction in CPU time in the batch window which provided additional head-room to stay within current LPAR caps and prevent increasing capacity on demand which would result in higher costs. Not to mention that critical jobs completed on time without impacting other workloads. They also benefited from a reduction in software costs due to a decrease in the MSUs consumed by sorting operations in the 4HRA peak window, and also are confident they can delay a processor upgrade which they saw as inevitable prior to deploying ZPSaver. Precisely ZPSaver is saving them money within their IT budget while improving operational efficiency. The mainframe team stands out in their IT organization for leveraging ZPSaver’s ready-to-implement innovations to address operational challenges and provide capacity for new opportunities.
The CIO of a US credit card processor, delivered a mandate to her Systems Operations and Administration Division – drive down mainframe costs. She wanted to see it done via efficiencies such as zIIP offload, tuning, elimination of unnecessary workloads or moving z/OS workloads to distributed systems.
Luckily for the Sys Ops and Admin team who supported a sort-intensive batch process, Precisely had recently developed ZPSaver, a value-add to MFSort for offloading a very significant amount of sorting work to zIIPs. This solution was the only one that did not involve programming resources.
The team did have to do their due diligence to confirm the cost savings but a POC demonstrated an average savings of 35 MSUs per hour in the peak billing window.
A North American automotive equipment manufacturer had completed a successful Syncsort Optimize IMS (formerly DB/2) POC and a that implementation was underway when they heard about MFSort for z/OS & Syncsort ZPSaver from the Precisely team as another way to address their mainframe cost reduction goals.
The Systems Manager was tasked with reducing mainframe costs but didn’t have the programming resources to dedicate to tuning projects. When he heard MFSort & ZPSaver could transparently replace DFSORT, use significantly less CPU time and elapsed time and make use of under-utilized zIIP engines, he was in.
Based on the trusted relationship he had built with Precisely he moved ahead with licensing the MFSort & ZPSaver bundle after just a brief POC.
He was able to show an immediate reduction in monthly software license charges and capacity projections showed an expensive hardware upgrade could be pushed out by a year.