Mainframe Computing and IT Storage Blog

Mainframe and Storage: Highlights from IBM's Technical Edge Conference Agenda

wired — Thu, 09 May 2013 14:39:00 GMT

The IBM Edge conference is June 10-14 in Las Vegas, Although the conference is held in Las Vegas this year, the sessions should bring enough expert material to divert mainframe storage managers from the gaming tables. Although mainframe isn't the primary focus, the Technical Edge agenda offers considerable material that addresses mainframe storage. Here are some of the sessions that look particularly interesting:

IBM z/OS Storage Management Ecosystem Update—provides an overview of IBM’s strategy for managing the z/OS storage ecosystems. Learn how the strategic z/OS storage management product, OMEGAMON XE for Storage, fits into the strategy and how the rest of the Tivoli z/OS Storage management portfolio works together to address common z/OS storage tasks and problem resolution. The session also promises to include a short review of Tivoli z/VM storage products.
IBM z/OS + DS8K Synergy—offers an overview of the IBM DS8000 architecture and its deep integration with IBM z/OS to deliver performance, high availability, optimization, and manageability. No other storage system has the unique integration with the IBM mainframe, according to IBM. That becomes particularly apparent when you start looking at Flash storage.
Introduction to the TS7700 Virtualization Engine Grid for the Uninitiated or Those Needing a Refresher—intended to introduce attendees to the IBM TS7700 Virtualization Engine, a virtualized tape solution. This basic level discussion will start by explaining why tape virtualization came about, and then cover basic and advanced concepts, up through the latest enhancements. If you are not already familiar with virtual tape you don’t want to miss this.
Retention management is an increasingly important and tricky topic these days, particularly as it ties in with compliance and even legal ediscovery. Edge 2013 offers two complementary sessions on retention management.
Untangling Retention Management under DFSMSrmm—DFSMS Removable Media Manager (DFSMSrmm) delivers a wide variety of retention controls for the z/OS tape resources it manages. Often the variety of options and the differences in how they behave can be confusing. The session promises to explain how and why DFSMSrmm does what it does so attendees can make the best decisions for their environment.
DFSMSrmm Best Practices, Features and New Stuff—DFSMS Removable Media Manager (DFSMSrmm) is one of the most regularly enhanced tape management systems in the market. Often, however, new features are overlooked. This session offers an overview of key product features that every DFSMSrmm data center should be taking full advantage of to properly safeguard their environments. In addition, it will detail several little-known product capabilities that can streamline administrator efficiency, allowing more time for other management activities.

Overall, IBM Edge 2013 will offer over 140 storage sessions, over 50 PureSystems sessions, more than 50 client case studies, and sessions on big data and analytics along with a full cloud track. In June, I'll be attending IBM Edge 2013. Look for me in the Social Media Lounge at the conference and in the sessions. You can follow me on Twitter for conference updates @Writer1225. I'll be using hashtag #IBMEdge to post live Twitter comments from the conference.

NOTE: Full disclosure: this blogger’s trip to Edge 2013 and related DancingDinosaur posts are being underwritten by IBM. However, the choice of content, ideas, and opinions are my own. Hope to see you there!

About these ads

IBM FlashSystem Remakes Data Center Economics

wired — Fri, 12 Apr 2013 21:30:00 GMT

Yesterday IBM announced the IBM FlashSystem, to drive Flash technology further into the enterprise. The IBM FlashSystem is a line of all-Flash storage appliances based on technology IBM acquired from Texas Memory Systems.

Flash can shorten the response of servers and storage systems to data requests from milliseconds to microseconds – an order of magnitude improvement. And because it is all electronic—nothing mechanical involved—and being delivered cost-efficiently at even petabyte scale, it can remake data center economics, especially for transaction-intensive and IOPS-intensive situations.

For example, the IBM FlashSystem 820 is the size of a pizza box but 20x faster than spinning hard drives and can store up to 24 TB of data. At the high end, you can assemble a 1 PB FlashSystem that fits in one rack and delivers 22 million IOs per second (IOPS). IBM calculates you would need 630 racks of high capacity hard disk drives or 315 racks of performance optimized disk to generate an equal amount of IOPS.

Mainframe shops already are familiar with Flash mainly in the form of cache and SSD. The zEnterprise makes extensive use of cache to boost performance and to ensure reliability and availability. The DS8000 storage line has been SSD capable for several years. The IBM System Storage DS8870, for example, comes equipped with IBM POWER7- based controllers. In a tiered storage environment it can automatically optimize the use of each storage tier, particularly SSD and now Flash, through the free IBM Easy Tier capability.

The IBM FlashSystem changes data center economics. One cloud provider reported deploying 5TB in 3.5 inches of rack space compared to deploying 1300 hard disks to achieve 400k IOPS and it did so at one-tenth the cost. Overall, Wikibon reports an all Flash approach will lower total system costs by 30%; that’s $4.9 million for all flash compared to $7.1 million for hard disk. Specifically, it reduced software license costs 38%, required 17% few servers, and lowered environmental costs by 74% and operational support costs by 35%. At the same time it boosted storage utilization by 50% while reducing maintenance and simplifying management with corresponding labor savings.

For data center managers, this runs counter to everything they learned about the cost of storage. Traditional storage economics starts with the cost of hard disk storage being substantially less than the cost of SSD or Flash on a $/GB basis. Organizations could justify SSD, however, by using it in small amounts to tap its sizeable cost/IOPS advantage for IOPS-intensive workloads.

IBM reversed traditional storage economics with the new FlashSystem storage by adopting a different approach to understanding the storage investment. Forget about cost/GB; even forget about cost/IOPS. Instead, focus on a systems perspective by considering all the costs involved in the total solution, from energy consumption to hard disk failure to labor to the cost of server software licensing. Then factor in the economic benefits of handling more transactions faster, more responsive systems, faster analytics, and more.

As reported in PC World, Steve Mills, IBM Senior Vice President put it this way at the introduction: Right now, generic hard drives cost about $2 per gigabyte. An enterprise hard drive will cost about $4 per gigabyte, and a high-performance hard drive will run about $6 per gigabyte. If an organization stripes its data across more disks for better performance, the cost goes up to about $10 per gigabyte. In some cases, where performance is critical, hard-drive costs can skyrocket to $30 or $50 per gigabyte. A solid state disk from IBM runs about $10 per gigabyte and can be filled to capacity, so they actually are less expensive in many cases.

And Mills was only talking from the cost/GB perspective; when you take a full systems perspective Flash looks even better. Said Ambuj Goyal, General Manager, Systems Storage, IBM Systems & Technology Group in the announcement: “The economics and performance of Flash are at a point where the technology can have a revolutionary impact on enterprises, especially for transaction-intensive applications.” But this actually goes beyond just transactions. Also look at big data analytics workloads, technical computing, and any other IOPS-intensive work.

As far as z data centers go, the IBM FlashSystem appliances, aimed primarily at open systems, are off in the future. However, mainframe data centers can continue to leverage SSD and Flash as they have and even expand it since it is increasingly easier to justify the investment, especially with an IBM enterprise-class SSD running about $10 per gigabyte. IBM further extends the value of SSD/Flash through the use of Real-time Compression and thin provisioning, which stretches your bang for the buck. So, using your workloads as the guide start thinking about cost-effectively working more SSD/Flash into your data center to lower costs.

IBM XIV Storage for the zEnterprise

wired — Thu, 01 Mar 2012 15:55:00 GMT

Even with the latest rev, v3.1, IBM’s open XIV grid storage is not likely to replace DS8800 storage as its recommended zEnterprise storage. Still, with an InfiniBand backplane delivering 4x the performance, the inclusion of 3 TB drives, use of SSD caching (up to 6TB of fast-read SSD) to boost performance and reduce latency, and with 99.999% availability XIV boasts impressive storage chops that might indeed interest a mainframe shop, especially one that was moving toward hybrid computing.

Last week DancingDinosaur looked at how a GUI when applied to mainframe management streamlines administration, shortens the learning curve, and reduces costs. This clearly was the case for El Al Airlines, a z10 shop when it opted for GUI-based management tools from CA Technologies.

XIV sports an impressive GUI of its own, one that is being ported to other IBM storage products, including the Storwize v7000, SONAS (Scale Out Network Attached Storage), and even the DS8000 lineup. XIV, which takes maximum advantage of built-in automation, had been considered pretty admin-friendly from the start. The latest GUI enhancements, such as the ability to search for any element across all managed systems or a reportedly fast, simple way to apply filters, only makes it easier. Mainframe shops can access the DS8000 GUI through Tivoli.

As z shops move inexorably toward cloud computing, especially private clouds, and hybrid computing XIV storage should increasingly come onto mainframe managements’ radar screen. For starters its extensive autonomic operations takes much of the management labor completely out of the process, meaning it can handle the speed, scale, and complexity of cloud provisioning, especially if a goal is to move to user self-service. In addition, it achieves very attractive price/performance. Finally, XIV scales linearly, with processors added along with the storage capacity, and it achieves 99.999% availability.
The autonomic operation should be particularly attractive. It eliminates sprawl by handling configuration, placement, and provisioning to achieve optimum results. It also eliminates the problem of hot spots due to over-subscription. And it is energy efficient; IBM reports $6445 in annual energy costs (no further details provided). However, in a price/performance benchmark IBM published, XIV beat the Hitachi USP-V. It also beat the DS8800 by a wide margin. Mainframe data center managers, take note.

In terms of throughput as measured in megabits per second (MBPS), however, the DS 8800 was tops at nearly 10,000 MBPS. The Hitachi USP V came next at about 8400 MBPS followed by the XIV Gen 3 at just under 8000 MBPS.

XIV constitutes a true storage grid. It consists of 15 cells, each with CPU, disk, DRAM, and SSD connected via a dual InfiniBand fabric. Each cell also includes a UPS and service modules. In terms of compute power, each XIV rack contains 60 CPU cores and 360 GB DRAM. You can choose 1, 2, or 3TB drives, which allows for a compact footprint. Data is dispersed automatically across all nodes, eliminating the need for RAID groups, and it uses independent parallel caches, which eliminates the need for a global cache. A patented distributed pre-fetch keeps a hot working data set in cache while extreme fault isolation between nodes ensures availability. Finally, performance scales linearly; each module added brings its own CPU, DRAM, SSD, and fast fabric.

Each host, however, sees XIV only a single large elastic block system with 24 8Gbit FC ports and 22 1Gbit iSCSI ports. Performance is boosted with SSD caching. Mainframe shops are not likely to ditch their DS8800 storage for XIV anytime soon. Those shops moving to private clouds and especially Linux on z or Linux on zBX blades, however, may find the XIV ease of operation quite appealing. Also, shops running hybrid workloads can benefit from XIV. Mainframe shops will probably manage XIV through Tivoli.

IBM and VMware Team Up on Storage

wired — Thu, 17 Nov 2011 21:11:00 GMT

IBM has been virtualizing its systems for decades. VMware dominates virtualization in the x86 systems environment and made it a strategic force there. Now these two virtualization industry leaders are teaming up and putting it to work for storage systems. Virtualized storage is essential for optimizing virtualized environments and backing up virtualized systems.

To date, virtualization has tended to focus on servers, providing the foundation for large scale server consolidation. Storage virtualization, however, has seemed at best to be an afterthought. That is until the advent of cloud computing. In his newest book Cloud, Virtualization, and Data Storage Networking (CRC Press, 2012) author and storage industry analyst Greg Schulz restores storage virtualization to its rightful spot alongside server virtualization. “The importance of main or processor (server) memory and external storage is that virtual machines need memory to exist when active and a place on disk to reside when not in memory,” he writes. In short, server and storage virtualization go hand in hand.

This becomes most obvious when trying to take advantage of the mobility of VMs, which is one of the most powerful capabilities of the virtualized environment. Effective VM mobility requires the associated storage move too, and do so non-disruptively. This requires close cooperation between the storage and the hypervisor.

The teaming up of IBM and VMware bring together the broadest portfolio of proven virtualization solutions for maximum choice to meet most organization’s needs. As this IBM Red Book shows, IBM storage and VMware have been working together for a while. They also are forging deep integration between their products for enhanced features and usability. Another more recent IBM Red Book on storage and VMware ESX is here. Finally, they promise to deliver the comprehensive service and support needed to simplify changes. For technology users, the partnership promises to dramatically reduce the complexity of IT virtualization and, in the process, significantly lower IT costs while increasing IT flexibility and business agility.

This is not just a sales or marketing alliance. IBM and VMware have joined together to complete real technical work with tangible implications in terms of features and capabilities. Specifically, IBM has 40 products certified on the VMware Compatibility Guide for vSphere 5.0. Also, IBM has taken advantage of all of VMware’s storage APIs. For example, integrating with VMware’s Site Recovery Manager (SRM) for backup to the cloud, VMware’s VAAI for array integration, its VADP for data protection, VMware VASA for storage tiering and allocation, VMware management plug-in APIs for IBM-specific storage management in vSphere.

This teamwork is continuing. They should be formally certifying metro storage clustering through the vVMC API, are engaging in active development of virtual volumes using VVOL, and using the VASA API in conjunction with VVOL for next generation storage awareness.

IBM also is leveraging vMotion with SVC to create stretched clusters up to 300 km apart. This will allow virtual servers and their storage to be transparently moved across data centers while SVC enables support for virtually all heterogeneous vendor storage.

The two companies also are using VMware’s SRM with IBM’s XIV to consolidate, ensure availability, and simplify disaster recovery of SAP deployments. They also have combined VMware View with IBM Storwize v7000 to enable VDI while using VMware View Composer to reduce storage capacity requirements by over 60%

To date, IBM reports more than 20,000 joint VMware customers. More importantly, given that storage and server virtualization are still evolving, the companies have collected over 800 VMware-IBM storage configurations in a reference database, which will prove invaluable as organizations look for what really works.

And IBM doesn’t expect to limit the experience it has gained with VMware to just that hypervisor. The storage requirements for virtual servers are not very different from one hypervisor to another. Expect to see similar things with IBM storage and virtualized servers running KVM or Hyper-V or maybe even Ubuntu.

Hot Summer for System z Storage

wired — Mon, 26 Sep 2011 15:22:00 GMT

Storage will become increasingly important in mainframe shops, especially as they carve out a role in virtualized, private clouds, and August will go down as a hot month for mainframe storage. EMC kicked off August by announcing a new virtual tape library for IBM z/OS environments followed a few days later by announcing new virtual storage capabilities for its Symmetrix VMAX for z/VM environments.

IBM jumped in just this past week by revealing various storage-related research initiatives for archiving, scale out file systems, storage clouds, SSD-based systems, and more. The company, however, has been introducing new storage, much of it applicable to the mainframe, all year.

EMC clearly plans to target IBM in the mainframe storage market. Early in August it introduced a new mainframe disk library it described as the first integrated solution for all tape use cases. The new disk library takes advantage of features in Data Domain, EMC’s deduplication line, and EMC VNX (unified storage for block, file, and object-based data).

The disk library, the DLm6000, enables mainframe users to minimize their storage and replication costs and dramatically improve their disaster recovery capabilities. EMC says it will address the full range of mainframe tape workloads with a single, consolidated all-disk system. By matching different workloads to the most appropriate storage, the DLm6000 also promises to maximize system performance and accelerate data retrievals and backup and recovery times.

EMC also announced new software for its Symmetrix VMAX family of storage systems, which will enhance virtual data center and cloud storage capabilities. Of particular interest is the new version of EMC AutoSwap software, which now supports z/VM environments, effectively expanding Symmetrix VMAX-based solutions for all of the major virtual machine operating systems. The new EMC GDDR (Geographically Dispersed Disaster Restart) for z/VM allows continuous availability and improved protection for virtualized mainframe environments.

IBM hasn’t exactly been twiddling its thumbs when it comes to enterprise storage. It enhanced DS8700 and DS8800 storage systems as well as SAN Volume Controller (SVC). At a recent storage research briefing IBM identified initiatives in both bit and logical data preservation and other enhancements to archiving. Other initiatives are looking at data density and data retention. A 50-year retention period is the latest goal; 100 years shouldn’t be far off. Similarly, IBM is aiming to pack 1 petabyte (PB) of data into a 1u slot; guess how much fits in a standard 40u rack. Behind much of IBM’s latest storage initiatives is its General Parallel File System (GPFS)

Maybe the most interesting IBM storage announcement this summer was XIV Gen 3. XIV focuses on storage for distributed systems but connects to the System z through z/VM via FC switches or SVC. XIV storage also can be accessed by z/VM guests through guest-attached FCP subchannels provided the applicable FCP multi-path driver is available. XIV should be of most interest with a heterogeneous Linux environment including Linux on z.

XIV may be IBM’s most innovative storage. It’s fully autonomic, meaning it can pretty much run itself. It is based on a grid design that connects a set of modules, each containing a powerful processor, memory, and storage. Multiple modules connect to create a seamless scalable storage grid. This design delivers predictable, sustained high performance storage with little or no intervention on the part the staff. Plus, it brings a slew of high reliability and availability capabilities. It also scales in near linear fashion: adding an XIV module increases storage capacity along with more CPU and RAM. Automatic rebalancing ensures load balance is maintained regardless of adding, deleting, or resizing volumes after new disk/module additions and even after a system component failure or during rebuild.

Watch DancingDinosaur for more enterprise storage info in the coming months.

Alan Radding, the author of DancingDinosaur, is a 20-year IT industry analyst and journalist covering mainframe, midrange, PC, web, and cloud computing. Feel welcome to check out his website -- http://www.technologywriter.com.