On 17 July 2017 IBM announced the z14 server as “the next generation of the world’s most powerful transaction system, capable of running more than 12 billion encrypted transactions per day. The new system also introduces a breakthrough encryption engine that, for the first time, makes it possible to pervasively encrypt data associated with any application, cloud service or database all the time”.
At first glance, a cursory review of the z14 announcement might just appear as another server upgrade release, but that could be a costly mistake by the reader. There are always subtle nuisances in any technology announcement, while finding them and applying them to your own business can sometimes be a challenge. In this particular instance, perhaps one might consider “Persuasive Encryption & Contained Pricing”…
When IBM releases a new generation of z Systems server, many of us look to the “feeds and speeds” data and ponder how that might influence our performance and capacity profiles. IBM state the average z14 speed compared with a z13 increase by ~10% for 6-way servers and larger. As per usual, there are software Technology Transition Offering (TTO) discounts ranging from 6% to 21% for z14 only sites. However, in these times where workload profiles are rapidly changing and evolving, it’s sometimes easy to overlook that IBM have to consider the holistic position of the IBM Z world. Quite simply, IBM has many divisions, Hardware, Software, Services, et al. Therefore there has to be interaction between the hardware and software divisions and in this instance, IBM have delivered a z14 server that is security focussed, with their Pervasive Encryption functionality.
Pervasive Encryption provides a simple and transparent approach for z Systems security, enabling the highest levels of data encryption for all data usage scenarios, for example:
- Processing: When retrieved from files and processed by applications
- In Flight: When being transmitted over internal and external networks
- At Rest: When stored in database structures or files
- In Store: When stored in magnetic storage media
Pervasive Encryption simplifies and reduces costs associated when protecting data by policy (I.E. Subset) or En Masse (I.E. All Of The Data, All Of the Time), achieving compliance mandates. When considering the EU GDPR (European Union General Data Protection Regulation) compliance mandate, companies must notify relevant parties within 72 hours of first having become aware of a personal data breach. Additionally organizations can be fined up to 4% of annual global turnover or €20 Million (whichever is greater), for any GDPR breach unless they can demonstrate that data was encrypted and keys were protected.
To facilitate this new approach for encryption, the IBM z14 infrastructure incorporates several new capabilities integrated throughout the technology stack, including Hardware, Operating System and Middleware. Integrated CPU chip cryptographic acceleration is enhanced, delivering ~600% increased performance when compared with its z13 predecessor and ~20 times faster than competitive server platforms. File and data set encryption is optimized within the Operating Systems (I.E. z/OS), safeguarding transparent and optimized encryption, not impacting application functionality or performance. Middleware software subsystems including DB2 and IMS leverage from these Pervasive Encryption techniques, safeguarding that High Availability databases can be transitioned to full encryption without stopping the database, application or subsystem.
Arguably IBM had to deliver this type of security functionality for its top tier z Systems customers, as inevitably they would be impacted by compliance mandates such as GDPR. Conversely, the opportunity to address the majority of external hacking scenarios with one common approach is an attractive proposition. However, as always, the devil is always in the detail, and given an impending deadline date of May 2018 for GDPR compliance, I wonder how many z Systems customers could implement the requisite z14 hardware and related Operating System (I.E. z/OS) and Subsystem (I.E. CICS, DB2, IMS, MQ, et al) .upgrades before this date? From a bigger picture viewpoint, Pervasive Encryption does offer the requisite functionality to apply a generic end-to-end process for securing all data, especially Mission Critical data…
Previously we have considered the complexity of IBM z Systems pricing mechanisms and in theory, the z14 announcement tried to simplify some of these challenges by building upon and formalizing Container Pricing. Container Pricing is intended to greatly simplify software pricing for qualified collocated workloads, whether collocated with other existing workloads on the same LPAR, deployed in a separate LPAR or across multiple LPARs. Container pricing allows the specified workload to be separately priced based on a variety of metrics. New approved z/OS workloads can be deployed collocated with other sub-capacity products (I.E. CICS, DB2, IMS, MQ, z/OS) without impacting cost profiles of existing workloads.
As per most new IBM z Systems pricing mechanisms of late, there is a commercial collaboration and exchange required between IBM and their customer. Once a Container Pricing solution is agreed between IBM and their customer, for an agreed price, an IBM Sales order is initiated, triggering the creation of an Approved Solution ID. The IBM provided solution ID is a 64-character string representing an approved workload with an entitled MSU capacity, representing a Full Capacity Pricing Container used for billing purposes.
Previously we considered the importance of WLM for managing z/OS workloads and its interaction with soft-capping, and this is reinforced with this latest IBM Container Pricing mechanism. The z/OS Workload Manager (WLM) enables Container Pricing using a resource classified as the Tenant Resource Group (TRG), defining the workload in terms of address spaces and independent enclaves. The TRG, combined with a unique Approved Solution ID, represents the IBM approved solution. As per standard SCRT processing, workload instrumentation data is collected, safeguarding that this workload profile does not directly impact the traditional peak LPAR Rolling Four-Hour Average (R4HA). The TRG also allows the workload to be metered and optionally capped, independent of other workloads that are running collocated in the LPAR.
MSU utilization of the defined workload is recorded by WLM and RMF, subsequently processed by SCRT to subtract the solution MSU capacity from the LPAR R4HA. The solution can then be priced independently, based on MSU resource consumed by the workload, or based upon other non-MSU values, specifically a Business Value Metric (E.g. Number of Payments). Therefore Container Pricing is much simpler and much more flexible than previous IBM collocated workload mechanism, namely IWP and zCAP.
Container Pricing eliminates the requirement to commission specific new environments to optimize MLC pricing. By deploying a standard IBM process framework, new workloads can be commissioned without impacting the R4HA of collocated workloads, being deployed as per business requirements, whether on the same LPAR, a separate LPAR, or dispersed across multiple LPAR structures. Quite simply, the standard IBM process framework is the Approved Solution ID, associating the client based z/OS system environment to the associated IBM sales contract.
In this first iteration release associated with the z14 announcement, Container Pricing can be deployed in the following three solution based scenarios:
- Application Development and Test Solution: Add up to 3 times more capacity to existing Development and Test environments without any additional monthly licensing costs; or create new LPAR environments with competitive pricing.
- New Application Solution: Add new z/OS microservices or applications, priced individually without impacting the cost of other workloads on the same system.
- Payments Pricing Solution: A single agreed value based price for software plus hardware or just software, via a number of payments processed metric, based on IBM Financial Transaction Manager (FTM) software.
IBM state z14 support for a maximum 2 million Docker containers in an associated maximum 32 TB memory configuration. In conjunction with other I/O enhancements, IBM state a z14 performance increase of ~300%, when compared with its z13 predecessor. Historically the IBM Z platform was never envisaged as being the ideal container platform. However, its ability to seamlessly support z/OS and Linux, while the majority of mission critical Systems Of Record (SOR) data resides on IBM Z platforms, might just be a compelling case for microservices to be processed on the IBM Z platform, minimizing any data latency transfer.
Container Pricing for z/OS is somewhat analogous to the IBM Cloud Managed Services on z Systems pricing model (I.E. CPU consumption based). Therefore, if monthly R4HA peak processing is driven by an OLTP application, or any other workload for that matter, any additional unused capacity in that specific SCRT reporting month can be allocated for no cost to other workloads. Therefore z/OS customers will be able to take advantage of this approach, processing collocated microservices or applications for a zero or nominal cost.
County Multiplex Pricing (CMP) Observation: The z14 is the first new generation of IBM Z hardware since the introduction of the CMP pricing mechanism. When a client first implements a Multiplex, IBM Z server eligibility cannot be older than two generations (I.E. N-2) prior to the most recently available server (I.E. N). Therefore the General Availability (GA) of z14, classifies the z114 and z196 servers as previously eligible CMP machines. IBM will provide a 3 Month grace period for CMP transition activities for these N-3 servers, namely z114 and z196. Quite simply, the first client CMP invoice must be submitted within 90 days of the z14 GA date, namely 13 September 2017, no later than 1 January 2018.
In conclusion, Pervasive Encryption is an omnipresent z14 function integrated into every data lifecycle stage, which could easily be classified as Persuasive Encryption, simplifying the sometimes arduous process of classifying and managing mission-critical data. As cybersecurity becomes an omnipresent clear and present danger, associated with impending and increasingly punitive compliance mandates such as GDPR, the realm of possibility exists to resolve this high profile corporate challenge once and for all.
Likewise, Container Pricing provides a much needed simple-to-use framework to drive MSU cost optimization for new workloads and could easily be classified as Contained Pricing. The committed IBM Mainframe customer will upgrade their z13 server environment to z14, as part of their periodic technology refresh approach. Arguably, those Mainframe customers who have been somewhat hesitant in upgrading from older technology Mainframe servers, might just have a compelling reason to upgrade their environments to z14, safeguarding cybersecurity challenges and evolving processes to contain z/OS MLC costs.