Why Deploy Your Cloud on z Systems

Why Deploy Your Cloud on z Systems

When most people think about cloud today, they start with the giant Cloud Service Providers – Amazon, Google, Azure, SalesForce, etc.  In many cases, these CSPs build their own (x86) servers, have proprietary APIs, and compete mostly on being a low cost provider.  Increasingly, the notion of Service Levels are being brought up along with the need to provide data privacy for the clients they are hosting.

What this paper is designed to do is to take a closer look at how some of the cloud capabilities which are available today on Z systems – with the recent enhancements of the new z13 system – can provide next generation of CSP’s competitive advantage in a rapidly growing segment of the market which requires higher QOS than commodity servers and ability to meet both industry and government regulatory compliance.

Why Cloud on Z

The mainframe has a long and well deserved reputation as being the most secure, highly available and scalable platform in the industry.  Almost all industries and large enterprises depend on z Systems to keep their applications continuously available and their data protected.  Part of this is due to the platform architecture which encompasses the processor, the firmware (i.e. hypervisor), much of the middlware and in the case of zOS, the operating system.  When we expand the aperture to Linux on z, we include the z/VM hypervisor and its secure isolation (e.g. EAL5+ certified).

In terms of QOS, zOS has capabilities like Workload Management (WLM) that enables us to run at near 100% utilization and manage stringent SLA’s based on business priorities.  For security, we have RACF for fine-grained controls with auditability and hardware encryption that is tamper proof.  We also have chargeback that can allocate costs down to the department/user level to provide the greatest degree of cost controls for applications that are being run.  Then, to top it off, Disaster Recovery solutions that are unmatched in the industry (i.e. GDPS – Geographically Dispersed Parallel Sysplex with data replication that is stateful).

Now, combine that with the ability to create and host services which can leverage existing assets with minimal to no change (e.g. zOS Connect) and a platform that can react dynamically and apply analytics to your System of Record Data with unmatched efficiency (i.e. DB2 w/IDAA) and you can see how “cloudifying” your zOS system offers existing enterprises to become Service Providers, both internally to their users as well as others in the industry for which they can offer differentiated services.

When we expand to look at Linux on z and the value that existing z Systems customers are realizing business value for a variety of workloads, it starts with TCO.  The ability to consolidate x86 workloads from a range of 6:1 to upwards of 20-30:1, based on the platforms processor speed, I/O bandwidth, memory overcommit (in z/VM), you have a compelling value proposition that has fueled double digit growth over the last 15 years.

Add to that the introduction of a whole new portfolio of open standards and open source environments, languages, tools, etc. such as OpenStack, Node.js, etc. and you have a platform that is both modern and relevant along with having economic factors that are compelling.    The OpenStack support includes rich set of Chef recipes enabled for Linux on Z workloads that can be integrated and orchestrated via OpenStack HEAT.  Plus, looking ahead to support for the KVM hypervisor which enables standardization on virtualization layer across platform and Docker containers which allows flexible deployment in consistent and easy way across platforms and you can support key development and deployment options with the highest QOS on any platform for mission critical workloads.

Looking forward, we think people will start looking to automate connections from their Systems of Record to their Systems of Engagement (e.g. for deploying new mobile applications).  By “service-enabling” their SOR (e.g. via zOS Connect) and deploying applications (images today, containers tomorrow) for SOE, you can have the best of both worlds.  And, in the case of our new Z Systems, you can have it all in one package (with high bandwidth/low-latency and secure connectivity)!

For Service Providers, the unique value the z platform brings to running Enterprise-grade Linux on a secure and efficient hypervisor (z/VM today, KVM tomorrow) combined with the enhanced reliability and on-demand capacity means reduced risk and competitive differentiation for a whole set of workloads that are looking for better data protection and ability to consistently meet stringent SLAs and comply with industry/government regulations.

Who says you can’t have your cake and eat it too!

Editors Note: This blog post was written by Mike Baskey who is an IBM Distinguished Engineer responsible for Cloud and the Mainframe.  Mike is relatively new to Twitter but can be followed at @MBaskey so please follow up via Twitter.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.