2018 Mainframe Solutions Edition - SE24
IBM z14 Features
he new IBM z14* system is designed for today's trust
economy. It provides clients with a robust platform for their
"Trust allows companies to take a leap in innovation, bring new
solutions to the market and use technology in new ways," says Tarun
Chopra, IBM Z* hardware offering manager. "Trust is what gives
the customers the confidence to share their personal information
and ultimately do business with your organization." (See "IBM z14
Features Improved Security, Insights and a Connected Ecosystem,"
The IBM z14 pervasive encryption keeps client data secure (see
"Pervasive Encryption Keeps All Data Secure" on page 27 in this
issue). The new platform also benefits cloud-based blockchain
offerings. By employing APIs, clients can link core assets to
IBM z14 performance enhancements help clients leverage the
trust economy. The system offers up to 170 configurable cores, or
up to 35 percent more total capacity than the largest z13*. It also
has up to 32TB of memory and a fault-tolerant redundant array of
independent memory designed to support memory availability. By
redesigning the L1 and L2 cache architecture, the z14 system has
1.5x more on-chip cache per core, compared with its predecessor.
Compression and crypto coprocessor performance also is
enhanced. Based on preliminary internal IBM lab measures, the
z14 has 6x faster encryption for like modes and data sizes with
enhanced on-chip cryptographic performance. It features the
new Crypto Express6S adapter installed in the PCIe I/O drawer
configuration and offers a tamper-resistant crypto coprocessor.
In SSL transactions, the Crypto Express6S
shows a 2x performance improvement.
To assist with innovation, the z14 has
24 // 2018 Learn more at: ibmsystemsmag.com/buyersguide
optimized simultaneous multithreading for Linux* performance.
"IBM has put its heart and soul into this," Chopra says.
Turning to workload performance issues, Stefan Reimbold,
development engineer at the IBM Research and Development
Lab in Boeblingen, Germany, describes reasons to collect and
analyze performance data on systems running Linux. Regular
monitoring and analysis of performance data will enable IT to more
easily investigate issues such as outages. (See "Linux Performance
Tools at a Glance," bit.ly/2lFlrdL.)
Unfortunately, most people don't collect data until a performance
issue occurs. "Performance analysis is done by comparing data of
a system in a good versus bad state. This helps to find out what
has changed and the cause of the change," he says.
IT can prepare for future performance issue analysis by setting
up regular monitoring and keeping data on hand since the last
major change. To analyze changes over time, historical data should
be kept on hand. The SYSSTAT utilities package has a key tool
for monitoring that can be used for regular monitoring as well as
data gathering, Reimbold says.
In addition to having data to analyze, IT must clearly define
the problem. Perhaps it occurs at a certain time of day or after
Once the data is in hand, it's time to employ tools for analysis.
Reimbold suggests a multistaged approach, as different tools
focus on specific analysis areas.
When a problem arises, IT must act quickly. Preparation, historical
data and making a clear statement about
the problem will aid in effective handling
of performance issues.