ATA DRIVEN GLOBAL VISION CLOUD PLATFORM STRATEG N POWERFUL RELEVANT PERFORMANCE SOLUTION CLO IRTUAL BIG DATA SOLUTION ROI FLEXIBLE DATA DRIVEN V

Transcription

1 ATA DRIVEN GLOBAL VISION CLOUD PLATFORM STRATEG N POWERFUL RELEVANT PERFORMANCE SOLUTION CLO IRTUAL BIG DATA SOLUTION ROI FLEXIBLE DATA DRIVEN V WHITE PAPER Build the Optimal Mainframe Storage Architecture With Hitachi Data Systems and Brocade Why Choose an IBM FICON Switched Network? By Bill Martin, Hitachi Data Systems Stephen Guendert, PhD, Brocade October 2016

2 WHITE PAPER 2 Contents Executive Summary 3 Introduction 4 Why Networked FICON Storage Is Better Than Direct-Attached Storage 4 Hitachi Virtual Storage Platform: Models VSP G1x00 and VSP F Why Brocade Gen5 DCX 8510 Is the Best FICON Director 4 An Ideal Pairing: Hitachi Virtual Storage Platform Models VSP G1x00 and VSP F1500 with Brocade Gen5 DCX Technical Reasons for a Switched FICON Architecture 5 Business Reasons for a Switched FICON Architecture 10 Why Switched FICON: Summary 15 Hitachi Virtual Storage Platform: Models VSP G1x00 and VSP F Scalability 16 Performance 16 IBM 3390 and FICON Support 16 Hitachi Dynamic Provisioning 17 Hitachi Dynamic Tiering 17 Hitachi Remote Replication 18 Multiplatform Support 20 Cost-Savings Efficiencies 21 Brocade Gen5 DCX 8510 in Mainframe Environments 21 Reliability, Availability and Serviceability 21 Proactive Performance Management 21 Scalability 22 Pair the Two Platforms Together 23 Traditional (z/os) Mainframe Environments 23 Linux on the Mainframe 23 FICON and FCP Intermix 23 Private Cloud 23 Conclusion 24

3 WHITE PAPER 3 Build the Optimal Mainframe Storage Architecture With Hitachi Data Systems and Brocade Executive Summary The IBM System z and newer zenterprise or, in other words, mainframes, continue to be a critical foundation in the IT infrastructure of many large companies today. An important element of the mainframe environment is the disk storage system (subsystem) that is connected to the mainframe via channels. The overall reliability, availability and performance of mainframe-based applications are dependent on this storage system. The performance demands, capacity, reliability, flexibility, efficiency and cost-effectiveness of the storage system are important aspects of any storage acquisition and configuration decision. The increasing demands for improved performance, that is, throughput (IOPS) and response time, make the storage system a critical element of the IT infrastructure. Another key factor in configuring the storage system is the decision of how it should be connected to the mainframe channels: direct attached to an IBM FICON channel or through a switched FICON network. This decision impacts the flexibility, reliability and availability of the storage infrastructure and the efficiency of the storage administrators. Hitachi Virtual Storage Platform models VSP G1x00 and VSP F1500 are all high-performance enterprise-class storage systems that provide a comprehensive set of storage and data services. These systems provide mainframe users with a cost-effective, highly reliable and available storage platform that delivers outstanding performance, capacity and scalability. VSP G1x00 and VSP F1500 support the operating systems used with IBM zenterprise processors, including z/os, z/vse, z/vm, and Linux on System z. This industry-leading storage series provides IBM 3390 disk drive support across a variety of disk drive types to meet the variety of performance and capacity needs of mainframe environments. The platform provides an internal physical disk capacity of approximately 8PB per storage system. With externally attached storage, VSP G1x00 and VSP F1500 can support up to 255PB of storage capacity. They also support 8Gb/sec FICON across all front-end ports for connectivity to the mainframe and 8Gb/sec Fibre Channel for connecting external storage. Using a FICON network configured with a switch or director to connect a storage system to the mainframe channels can significantly enhance reliability, flexibility and availability of storage systems. At the same time, it can maximize storage performance and throughput. A switched FICON network allows the implementation of a fan-in, fan-out configuration, which allows maximum resource utilization and simultaneously helps localize failures, improving availability. The Brocade Gen5 DCX 8510 is a backbone-class FICON or Fibre Channel director. The Brocade Gen5 DCX 8510 family of FICON directors provides the industry's most powerful switching infrastructure for modern mainframe environments. It provides the most reliable, scalable, efficient, cost-effective, high-performance foundation for today's highly virtualized mainframe environments. The Brocade Gen5 DCX 8510 builds upon years of innovation and experience and leverages the core technology of Brocade systems, providing over % uptime in the world's most demanding data centers. The Gen5 DCX 8510 supports the operating systems used with zenterprise processors: z/os, z/vse, z/vm, Linux on System z, and ztpf for System z. This industry-leading FICON director supports 2, 4, 8, 10, and 16Gb/sec Fibre Channel links, FICON I/O traffic, and 1 gigabit Ethernet (GbE) or 10GbE links on Fibre Channel over IP (FCIP). At the same time, it provides 8.2Tb/sec chassis bandwidth. The combination of switched FICON connectivity with Hitachi s VSP G1x00 and VSP F1500 connected to mainframe channels through a Brocade Gen5 DCX 8510 Director provides a powerful, flexible and highly available solution. Together, they support the storage features, performance and capacity needed for today s mainframe environments.

4 WHITE PAPER 4 Introduction This paper explores both technical and business reasons for implementing a switched FICON architecture instead of a direct-attached storage FICON architecture. It also explains why Hitachi Virtual Storage Platform (VSP) models VSP G1x00 and VSP F1500 with the Brocade FICON Director together provide an outstanding, industry-leading solution for FICON environments. With the many enhancements and improvements in mainframe I/O technology in the past five years, the question "Do I need FICON switching technology, or should I go with direct-attached storage?" is frequently asked. With up to 320 FICON Express8S channels supported on an IBM zenterprise z114, z196 and zec12 and zbc12, why not just direct-attach the control units? The short answer is that with all of the I/O improvements, switching technology is needed now more than ever. In fact, there are more reasons to use switched FICON than there were to use switched IBM ESCON. Some of these reasons are purely technical; others are more business-related. Why Networked FICON Storage Is Better Than Direct-Attached Storage The raw bandwidth of FICON Express8S running on IBM zenterprise systems is 40 times greater than the capabilities of ESCON. The raw I/Os per second (IOPS) capacity of FICON Express8S channels is even more impressive, particularly when a channel program uses the z High Performance FICON (zhpf) protocol. To utilize these tremendous improvements, the FICON protocol is packet-switched and, unlike ESCON, capable of having multiple I/Os occupy the same channel, simultaneously. FICON Express8S channels on zenterprise processors can have up to 64 concurrent I/Os (open exchanges) to different devices. FICON Express8S channels running zhpf can have up to 750 concurrent I/Os on the zenterprise processor family. Only when a director or switch is used between the host and storage device can the true performance potential of channel bandwidth and I/O processing gains be fully exploited. Hitachi Virtual Storage Platform: Models VSP G1x00 and VSP F1500 Hitachi Virtual Storage Platform models VSP G1x00 and VSP F1500, with their vast functionality and throughput capability, are ideal for IBM mainframe environments and provides a comprehensive set of storage and data services. The flexibility in configuring, partitioning and tiering storage, VSP G1x00 and VSP F1500 easily support mainframe environments, with multiple LPARS running multiple operating system images in the same sysplex. The packaging, enhanced features and improved manageability of VSP G1x00 and VSP F1500 provide mainframe users with a cost-effective, highly reliable and available storage platform. It delivers outstanding performance, capacity and scalability. The storage platform easily supports both mainframe and open systems environments. For mainframe environments, it supports z/os, z/vse, z/vm and ztpf. Many organizations are considering the benefits of running Linux on IBM zenterprise processors. VSP G1x00 and F1500 support this capability for both count key device (CKD) and fixedblock architecture (FBA) disk formats and provide a solid foundation for implementing private clouds. With support for FICON Express8S and the support of 2Gb, 4Gb and 8Gb FICON and 2Gb, 4Gb and 8Gb Fibre Channel connectivity, this platform delivers industry-leading I/O performance. VSP G1x00 and VSP F1500 can have up to 24 front-end directors with a total of 176 FICON ports. Each port can support more IOPS than a single zenterprise FICON Express8 channel can deliver. As a result, it is ideally suited for connectivity to the mainframe through a switched FICON network. Why Brocade Gen5 DCX 8510 Is the Best FICON Director Emerging and evolving enterprise-critical workloads and higher-density virtualization are continuing to push the limits of SAN infrastructures. This is even more true in a data center with IBM zenterprise and its support for

5 WHITE PAPER 5 Microsoft Windows in the zenterprise Blade Center Extension (zbx). The Brocade Gen5 DCX 8510 family features industry-leading 16Gb/sec performance, and 8.2Tb chassis bandwidth to address these next-generation I/O and bandwidth-intensive application requirements. In addition, the Brocade Gen5 DCX 8510 provides unmatched slot-toslot and port performance, with 512Gb/sec bandwidth per slot (port card or blade). And this performance comes in the most energy-efficient FICON director in the industry, using an average of less than 1 watt per Gb/sec, which is 15 times more efficient than competitive offerings. The Brocade Gen5 DCX 8510 family enables high-speed replication and backup solutions over metro or WAN links with native Fibre Channel (10Gb/sec or 16Gb/sec). FCIP 1GbE or 10GbE extension support is optional. These solutions are accomplished by integrating this technology via a blade (FX24-8) or standalone switch (Brocade 7800). Brocade Fabric Vision technology, an extension of Generation 5 Fibre Channel, has been introduced and qualified on IBM System z with Brocade Fabric Operating System (FOS) 7.2. Fabric Vision provides a breakthrough hardware and software diagnostic, monitoring and management solution that unleashes the full potential of high-density server and storage virtualization, cloud architectures and next-generation storage. Finally, this solution is accomplished with unsurpassed levels of reliability, availability and serviceability (RAS), based upon more than 25 years of Brocade experience in the mainframe space. This experience includes defining FICON standards and authoring or co-authoring many FICON patents. An Ideal Pairing: Hitachi Virtual Storage Platform Models VSP G1x00 and VSP F1500 with Brocade Gen5 DCX 8510 The IBM zenterprise architecture is the highest performing, most scalable, cost-effective, energy-efficient platform in mainframe computing history. To get the most out of your investment in IBM zenterprise, you need a storage infrastructure, that is, a DASD platform and FICON director that can match the impressive capabilities of zenterprise. Hitachi Data Systems and Brocade, via VSP G1x00 and VSP F1500 and Gen5 DCX 8510, offer the highest performing and most reliable, scalable, cost-effective and energy-efficient products in the storage and networking industry. The experience of these two companies in the mainframe market, coupled with the capabilities of VSP G1x00 and VSP F1500, as well as Gen5 DCX 8510, make pairing them with IBM zenterprise the ideal best in industry storage architecture for mainframe data centers. Why IT Should Choose Networked Storage for FICON Over Direct-Attached Storage Direct-attached FICON storage might appear to be a great way to take advantage of FICON technology. However, a closer examination will show why a switched FICON architecture is a better, more robust design for enterprise data centers than direct-attached FICON. Technical Reasons for a Switched FICON Architecture There are five key technical reasons for connecting storage control units using switched FICON: Overcome buffer credit limitations on FICON Express8 channels. Build fan-in, fan-out architecture designs for maximizing resource utilization. Localize failures for improved availability. Increase scalability and enable flexible connectivity for continued growth. Leverage new FICON technologies.

6 WHITE PAPER 6 FICON Channel Buffer Credits When IBM introduced the availability of FICON Express8 channels, one very important change was the number of buffer credits available on each port per four-port FICON Express8 channel card. While FICON Express4 channels had 200 buffer credits per port on a four-port FICON Express4 channel card, this changed to 40 buffer credits per port on a FICON Express8 channel card. Organizations familiar with buffer credits will recall that the number of buffer credits required for a given distance varies directly in a linear relationship with link speed. In other words, doubling the link speed would double the number of buffer credits required to achieve the same performance at the same distance. Also, organizations might recall the IBM System z10 Statement of Direction concerning buffer credits: "The FICON Express4 features are intended to be the last features to support extended distance without performance degradation. IBM intends to not offer FICON features with buffer credits for performance at extended distances. Future FICON features are intended to support up to 10km without performance degradation. Extended distance solutions may include FICON directors or switches (for buffer credit provision) or dense wavelength division multiplexers [DWDM] (for buffer credit simulation)." IBM held true to its statement, and the 40 buffer credits per port on a FICON Express8/FICON Express8S channel card can support up to 10km of distance for full-frame size I/Os (2KB frames). What happens if an organization has I/Os with smaller than full-size frames? The distance supported by the 40 buffer credits would increase. It is likely that at faster future link speeds, the distance supported will decrease to 5km or less. A switched architecture allows organizations to overcome the buffer credit limitations on the FICON Express8/FICON Express8S channel card. Depending upon the specific model, FICON directors and switches can have more than 1300 buffer credits available per port for long-distance connectivity. Fan-In, Fan-Out Architecture Designs In the late 1990s, the open systems world started to implement Fibre Channel storage area networks (SANs) to overcome the low utilization of resources inherent in a direct-attached storage architecture. SANs addressed this issue through the use of fan-in and fan-out storage network designs. That is, multiple server host bus adapters (HBAs) could be connected through a Fibre Channel switch to a single storage port: in other words, fan-in. Or, a single-server HBA could be connected through a Fibre Channel switch to multiple storage ports: that is, fan-out. These same principles apply to a FICON storage network. As a general rule, FICON Express8 and FICON Express8S channels offer different levels of performance, in terms of IOPS and bandwidth, than the storage host adapter ports to which they are connected. Therefore, a direct-attached FICON storage architecture may see very low channel or storage port utilization rates. To overcome this issue, fan-in and fan-out storage network designs are used. A switched FICON architecture allows a single channel to fan out to multiple storage devices via switching, improving overall resource utilization. This capability can be especially valuable if an organization's environment has newer FICON channels, such as FICON Express8 or Express8S, but older tape drive technology. Figure 1 illustrates how a single FICON channel can concurrently keep several tape drives running at full-rated speeds. The actual fan-out ratios for connectivity to tape drives will, of course, depend on the specific tape drive and control unit; however, it is not unusual to see a FICON Express8 or Express8S channel fan-out from a switch to five to six tape drives (a 1:5 or 1:6 fan-out ratio). The same principles apply for fan-out to storage systems. The exact fan-out ratio is dependent on the storage system model and host adapter capabilities for IOPS and/or bandwidth. On the other hand, several FICON channels could be connected through a director or switch to a single storage port to maximize the port utilization and increase overall I/O efficiency and throughput.

7 WHITE PAPER 7 Figure 1. Switched FICON allows one channel to keep multiple tape drives fully utilized. Keep Failures Localized In a direct-attached architecture, a failure anywhere in the path renders both the channel interface and the control unit port unusable. The failure could be of: an entire FICON channel card, a port on the channel card, the cable, the entire storage host adapter card, or an individual port on the storage host adapter card. In other words, a failure on any of these components will affect both the mainframe connection and the storage connection. The worst possible reliability, availability and serviceability for FICON-attached storage are provided with a direct-attached architecture. With a switched architecture, failures are localized to only the affected FICON channel interface or control unit interface, not both. The nonfailing side remains available, and if the storage side has not failed, other FICON channels can still access that host adapter port via the switch or director (see Figure 2). This failure isolation, combined with fan-in and fan-out architectures, allows for the most robust storage architectures, minimizing downtime and maximizing availability.

8 WHITE PAPER 8 Figure 2. A FICON director isolates faults and improves availability. Scalable and Flexible Connectivity Direct-attached FICON does not easily allow for dynamic growth and scalability, since a single FICON channel card port is tied to a single dedicated storage host adapter port. In such an architecture, there is a 1:1 relationship (no fan-in or fan-out). Since there is a finite number of FICON channels available (dependent on the mainframe model or machine type), growth in a mainframe storage environment with such an architecture can pose a problem. What happens if an organization needs more FICON connectivity, but has run out of FICON channels? FICON switching and proper usage of fan-in and fan-out in the storage architecture design will go a long way toward improving scalability. In addition, best-practice storage architecture designs include room for growth. With a switched FICON architecture, adding a new storage system or port in a storage system is much easier: Simply connect the new storage system or port to the switch. This eliminates the need to open the channel cage in the mainframe to add new channel interfaces, reducing both capital and operational expenditures (capex and opex). This also gives managers more flexible planning options when upgrades are necessary, since the urgency of upgrades is lessened. What about the next generation of channels? The bandwidth capabilities of channels are growing at a much faster rate than those of storage devices. As channel speeds increase, switches will allow data center managers to take advantage of new technology as it becomes available, while protecting investments and minimizing costs. Also, it is an IBM best-practice recommendation to use single-mode long-wave connections for FICON channels. Storage vendors, however, often offer single-mode long-wave connections and multimode short-wave connections on their storage systems, allowing organizations to decide which to use. The organization makes the decision based on the trade-off between cost and reliability. Some organizations' existing storage devices have a mix of single-mode and multimode connections. Since they cannot directly connect a single-mode FICON channel to a

9 WHITE PAPER 9 multimode storage host adapter, this could pose a problem. With a FICON director or switch in the path, however, organizations do not need to change the storage host adapter ports to comply with the single-mode best-practice recommendation for the FICON channels. The FICON switching device can have both types of connectivity. It can have single-mode long-wave ports for attaching the FICON channels, and multimode short-wave ports for attaching the storage. Furthermore, FICON switching elements at 2 different locations can be interconnected by fiber at distances up to 100km or more, creating a cascaded FICON switched architecture. This setup is typically used in disaster recovery and business continuance architectures. As previously discussed, FICON switching allows resources to be shared. With cascaded FICON switching, those resources can be shared between geographically separated locations, allowing data to be replicated or tape backups to be made at the alternate site, away from the primary site, with no performance loss. Often, workloads will be distributed such that both the local and remote sites are primary production sites, and each site uses the other as its backup. While the fiber itself is relatively inexpensive, laying new fiber may require an expensive construction project. While DWDM can help get more out of fiber connections, inter-switch links with up to 16Gb/sec of bandwidth are offered by switch vendors and can reduce the cost of or even eliminate the need for DWDM. FICON switches maximize utilization of this valuable intersite fiber by allowing multiple environments to share the same fiber link. In addition, FICON switching devices offer unique storage network management features, such as ISL trunking and preferred pathing, which are not available with DWDM equipment. FICON switches allow data center managers to further exploit intersite fiber sharing by enabling them to intermix FICON and native Fibre Channel Protocol (FCP) traffic, which is known as Protocol Intermix Mode, or PIM. Even in data centers where there is enough fiber to separate FICON and open systems traffic, preferred pathing features on a FICON switch can be great cost savers. With preferred paths established, certain cross-site fiber can be allocated for the mainframe environment, while other fiber can be allocated for open systems. The ISLs can be configured such that in the event of a failure, and only in the event of an ISL failure, the links would be shared by both open systems and mainframe traffic. Leverage New Technologies Over the past five years, IBM has announced a series of technology enhancements that require the use of switched FICON. These include: N_port ID virtualization (NPIV) support for z Linux. FICON Dynamic Channel-Path Management (DCM). z/os FICON Discovery and Auto-Configuration (zdac). IBM announced support for NPIV on z Linux in Today, NPIV is supported on the System z9, z10, z196, z114, zec12 and zbc12. Until NPIV was supported on System z, adoption of Linux on System z had been relatively slow. NPIV allows for full support of LUN masking and zoning by virtualizing the Fibre Channel identifiers. This, in turn, allows each Linux on System z image to appear as if it has its own individual HBA when those images are, in fact, sharing FCP channels. Since IBM began supporting NPIV on System z, adoption of Linux on System z has grown significantly. IBM believes approximately 24% of MIPS shipping on new zenterprise processors are for Linux on System z implementations. Implementation of NPIV on System z requires a switched architecture. FICON DCM is another feature that requires a switched FICON architecture. FICON DCM provides the ability to have System z automatically manage FICON I/O paths connected to storage systems in response to changing workload demands. Use of FICON DCM helps simplify I/O configuration planning and definition, reduces the complexity of

10 WHITE PAPER 10 managing I/O, dynamically balances I/O channel resources, and enhances availability. FICON DCM can best be summarized as a feature that allows for more flexible channel configurations, by designating channels as "managed," and proactive performance management. FICON DCM requires a switched FICON architecture because topology information is communicated via the switch or director. The FICON switch must have a control unit port (CUP) license and be configured or defined as a control unit in the hardware configuration definition (HCD). z/os FICON Discovery and Auto-Configuration (zdac) is the latest technology enhancement for FICON. IBM introduced zdac as a follow-on to an earlier enhancement in which the FICON channels log into the Fibre Channel name server on a FICON director. zdac enables the automatic discovery and configuration of FICON-attached DASD and tape devices. Essentially, zdac automates a portion of the HCD Sysgen process. zdac uses intelligent analysis to help validate the System z and storage definitions' compatibility, and uses built-in best practices to help configure for high availability and avoid single points of failure. zdac is transparent to existing configurations and settings. It is invoked and integrated with the z/os HCD and z/os Hardware Configuration Manager (HCM). zdac also requires a switched FICON architecture. IBM also introduced support for transport-mode FICON (known as z High Performance FICON, or zhpf) in October 2008 and announced enhancements in July While not required for zhpf, a switched architecture is recommended. Business Reasons for a Switched FICON Architecture In addition to the technical reasons described earlier, the following business reasons support implementing a switched FICON architecture: Enable massive consolidation in order to reduce capex and opex. Improve application performance at long distances. Support growth and enable effective resource sharing. Massive Consolidation With NPIV support on System z, server and I/O consolidation is very compelling (see Figure 3). IBM undertook a well-publicized project at its internal data centers (Project Big Green) and consolidated 3900 open systems servers onto 30 System z mainframes running Linux. IBM's total cost of ownership (TCO) savings were calculated, taking into account footprint reductions, power and cooling, and management simplification costs. The result was nearly 80% TCO savings for a five-year period. This scale of TCO savings is why 24% of new IBM mainframe processor shipments are now being used for Linux. Implementation of NPIV requires connectivity from the FICON (FCP) channel to a switching device (director or smaller port-count switch) that supports NPIV. A special microcode load is installed on the FICON channel to enable it to function as an FCP channel. NPIV allows the consolidation of up to 255 z Linux images ("servers") behind each FCP channel, using one port on a channel card and one port on the attached switching device for connecting these virtual servers. This enables massive consolidation of many HBAs, each attached to its own switch port in the SAN. As a best practice, IBM currently recommends configuring no more than 32 Linux images per FCP channel. This level of I/O consolidation was possible prior to NPIV support on System z. However, implementing LUN masking and zoning in the same manner as with open systems servers, SAN and storage was not possible prior to the support for NPIV with Linux System z. NPIV implementation on System z has also resulted in consolidation and adoption of a common SAN for distributed or open systems (FCP) and mainframe (FICON), commonly known as protocol intermix mode (PIM). While IBM has

11 WHITE PAPER 11 supported PIM in System z environments since 2003, adoption rates were low until NPIV implementations for Linux for System Z picked up with the introduction of System z10 in With z10, enhanced segregation and security beyond simple zoning was possible through switch partitioning or virtual fabrics and logical switches. With 24% of new mainframes being shipped for use with Linux on System z, it is safe to say that at least 19% of mainframe environments are now running a shared PIM environment. Leveraging enhancements in switching technology, performance and management, PIM users can now fully populate the latest high-density directors with minimal or no oversubscription. They can use management capabilities, such as virtual fabrics or logical switches to fully isolate open systems ports and FICON ports in the same physical director chassis. Rather than having more partially populated switching platforms that are dedicated to either mainframe (FICON) or open systems (FCP), PIM allows for consolidation onto fewer physical switching devices. It reduces management complexity and improves resource utilization. This, in turn, leads to lower operating costs, and a lower TCO for the storage network. It also allows for a consolidated, simplified cabling infrastructure. Figure 3. Organizations implement NPIV to consolidate I/O in z Linux environments. Application Performance Over Distance As previously discussed, the number of buffer credits per port on a four-port FICON Express8 channel has been reduced to 40, supporting up to 10km without performance degradation. What happens if an organization needs to go beyond 10km for a direct-attached storage configuration? They will likely see performance degradation due to insufficient buffer credits. Without a sufficient quantity of buffer credits, the "pipe" cannot be kept full with streaming frames of data.

12 WHITE PAPER 12 Switched FICON avoids this problem (see Figure 4). FICON directors and switches have a sufficient quantity of buffer credits available on ports to allow them to stream frames at full-line performance rates with no bandwidth degradation. IT organizations that implement a cascaded FICON configuration between sites can, with the latest FICON director platforms, stream frames at 16Gb/sec rates. And they experience no performance degradation for sites that are 100km apart. Switched FICON technology also allows organizations to take advantage of hardware-based FICON protocol acceleration or emulation techniques for tape (reads and writes). This emulation technology is available on standalone extension switches or on a blade in FICON directors. It allows the z/os-initiated channel programs to be acknowledged locally at each site and avoids the back-and-forth protocol handshakes that normally travel between remote sites. It also reduces the impact of latency on application performance and delivers local-like performance over unlimited distances. In addition, this acceleration or emulation technology optimizes bandwidth utilization. Why is bandwidth efficiency so important? It is typically the most expensive budget component in an organization's multisite disaster recovery or business continuity architecture. Anything that can be done to improve the utilization and/or reduce the bandwidth requirements between sites would likely lead to significant TCO savings. Figure 4. Switched FICON with emulation allows optimized performance and bandwidth utilization over extended distance.

13 WHITE PAPER 13 Enable Growth and Resource Sharing Direct-attached storage forces a 1:1 relationship between host connectivity and storage connectivity. In other words, each storage port on a storage system host adapter requires its own physical port connection on a FICON Express8 channel card. These channel cards are typically very expensive on a per-port basis: typically four to six times the cost of a FICON director port. Also, there is a finite number of FICON Express8S channels available on a zenterprise (a maximum of 320), as well as a finite number of host adapter ports in the storage system. If an organization has a large configuration and a direct-attached FICON storage architecture, how does it plan to scale its environment? What happens if an organization acquires a company and needs additional channel ports? A switched FICON infrastructure allows cost-effective, seamless expansion to meet growth requirements. Direct-attached FICON storage also typically results in underutilized host channel card ports and host adapter ports in storage systems. FICON Express8 and FICON Express8S channels can comfortably perform at high-channel utilization rates, and a direct-attached storage architecture typically sees channel utilization rates of 10% or less. As illustrated in Figure 5, leveraging FICON directors or switches allows organizations to maximize channel utilization. Figure 5. Switched FICON drives improved channel utilization, while preserving CHPIDs for growth.

14 WHITE PAPER 14 It also is very important to keep traffic for tape drives streaming, and to avoid stopping and starting the tape drives. These actions lead to unwanted wear and tear of tape heads, cartridges and the tape media itself. Using FICON acceleration or emulation techniques, as described earlier, streaming can be accomplished with a configuration similar to the one shown in Figure 6. Such a configuration requires solid analysis and planning, but it will pay dividends for an organization's FICON tape environment. Figure 6. A well-planned configuration can maximize CHPID capacity utilization for FICON tape efficiency. Finally, switches facilitate fan-in, which allows different hosts and LPARs whose I/O subsystems are not shared to share the same assets. While some benefits may be realized immediately, the potential for value in future equipment planning can be even greater. With the ability to share assets, equipment that would be too expensive for a single environment can be deployed in a cost-saving manner. The most common example is to replace tape farms with virtual tape systems. By reducing the number of individual tape drives, maintenance (service contracts), floor space, power, tape handling and cooling costs are reduced. Virtual tape also improves reliable data recovery, allows for significantly shorter recovery time objectives (RTO) and nearer recovery point objectives (RPO), and offers features such as peer-to-peer copies. However, without the ability to share these systems, it may be difficult to amass sufficient cost savings to justify the initial cost of virtual tape. And the only practical way to share these standalone tape systems or tape libraries is through a switch. With disk storage systems, in addition to sharing the asset, it is sometimes desirable to share the data across multiple systems. The port limitations on a storage system may prohibit or limit this capability using direct-attached (point-to-point) FICON channels. Again, the switch can provide a solution to this issue. Even when there is no need to share devices during normal production, this capability can be very valuable in the event of a failure. Data sets stored on tape can quickly be read by CPUs picking up workload that is already attached to the same switch as the tape drives. Similarly, data stored on a storage system can be available as soon as a fault is determined. Switch features, such as preconfigured port prohibit or allow matrix tables, can ensure that access intended only for a disaster scenario is prohibited during normal production.

15 WHITE PAPER 15 Why Switched FICON: Summary Direct-attached FICON might appear to be a great way to take advantage of FICON technology's advances over ESCON. However, a closer examination shows that switched FICON, similar to switched ESCON, is a better, more robust architecture for enterprise data centers. Switched FICON offers: Better utilization of host channels and their performance capabilities. Scalability to meet growth requirements. Improved reliability, problem isolation and availability. Flexible connectivity to support evolving infrastructures. More robust business continuity implementations via cascaded FICON. Improved distance connectivity, with improved performance over extended distances. New mainframe I/O technology enhancements such as NPIV, FICON DCM, zdac and zhpf. Switched FICON also provides many business advantages and potential cost savings, including: The ability to perform massive server, I/O and SAN consolidation, dramatically reducing capex and opex. Local-like application performance over any distance, allowing host and storage resources to reside wherever business dictates. More effective resource sharing, improved utilization, reduced costs and improved recovery time. The trend toward increased usage of Linux on System z is growing, and there are cost advantages of NPIV implementations and PIM SAN architectures. As a result, direct-attached storage in a mainframe environment is becoming a thing of the past. Investments made in switches for disaster recovery and business continuance are likely to pay the largest dividends. Having access to alternative resources and multiple paths to those resources can result in significant savings in the event of a failure. The advantages of a switched FICON infrastructure are simply too great to ignore. Hitachi Virtual Storage Platform: Models VSP G1x00 and VSP F1500 Hitachi Data Systems has over 25 years of experience supporting IBM mainframe environments. A large portion of the installed base of Hitachi storage systems connects to IBM z/os and S/390 mainframes via ESCON and FICON networks. Hitachi Virtual Storage Platform models VSP G1x00 and VSP F1500 build on this experience and introduce new features and packaging to improve performance while lowering TCO. They feature lower power and cooling requirements, high-density packaging based on industry-standard 19-inch racks and faster microprocessors. They also offer the choice of disk drives types, including solid-state disk (SSD), Hitachi Accelerated Flash storage, serial attached SCSI (SAS) and nearline SAS. This storage platform provides an industry-leading, reliable and highly available storage system for mainframes in IBM z/os environments: It supports z/os, z/vse, z/vm and z/tpffor zenterprise. Many organizations are considering the benefits of or running Linux on IBM zenterprise processors. VSP G1x00 and VSP F1500 support this capability for both CKD and FBA disk formats and provides a solid foundation for implementing private clouds. Hitachi has implemented key performance features in support of these operating systems running on zenterprise, including PAV, HyperPAV, z/hpf, Multiple Allegiance, MIDAW and priority I/O queuing. It also provides a unique mainframe storage management solution to deliver functionally compatible extended address volumes (EAV) for z/os, data volume expansion (DVE), and IBM FlashCopy SE (with space efficiency capability).

16 WHITE PAPER 16 VSP G1x00 and VSP F1500 are designed to be highly available and resilient. All critical components are implemented in pairs. If a component fails, the paired component can take over the workload without an outage. With its support of multiple RAID configurations, an organization s data is protected in event of a disk drive problem. Additionally, VSP G1x00 and VSP F1500 offer industryleading replication software. Its support of FlashCopy and FlashCopy SE provides the functionality of IBM z/os Metro Mirror. And it replicates to remote locations using Hitachi Universal Replicator (HUR), allowing copies of data can be maintained locally and at remote locations. This practice ensures data availability in case the primary copy becomes unusable or is not accessible. Scalability Hitachi Virtual Storage Platform models VSP G1x00 and VSP F1500 can scale to provide increased performance, capacity, throughput and connectivity while dynamically combining multiple units into a single logical system with shared resources. They can also virtualize new and existing external storage systems. This scaling means that VSP G1x00 and VSP F1500 can grow nondisruptively to meet changing needs within the data center. They minimize outages to extend the platform and enhance functionality while providing flexibility in the configuration and choice of disk technology to meet the specific needs of each environment. VSP G1x00 and VSP F1500 controller-based storage virtualization enables connectivity to and virtualization of external storage, which can potentially extend the life of existing storage assets and reduce costs. Virtualizing external storage: Enables the reuse of existing or legacy assets for less critical or accessed data. Compatability With VSP G1x00 and VSP F1500 LEARN MORE Simplifies management of external storage with common management and data protection for internal and external storage. Supports the reuse of existing or legacy assets across data centers within a metro area network distance and across global distances with replication capabilities of the scale up storage system. Performance Hitachi Virtual Storage Platform models VSP G1x00 and VSP F1500 usher in a new level of I/O throughput, response and scalability. They support 8Gb and 16Gb/sec FICON (FICON Express8 and FICON Express8S). A single VSP G1x00 or VSP F1500 FICON 8Gb port can handle higher traffic rates, which can be delivered by a single zenterprise FICON Express8 or FICON Express8S channel. This storage networking is critical to optimizing performance and maximizing throughput in mainframe environments. IBM 3390 and FICON Support This industry-leading storage system provides 3390 disk drive support through emulation across a variety of disk drive types to meet the variety of performance and capacity needs of mainframe environments. The platforms support SSD flash drives and Hitachi Acdelerated Flash modules, providing ultra-high-speed, as well as 2.5-inch SAS drives and nearline SAS drives. They can control up to 65,280 logical volumes and provide an internal physical disk capacity of approximately 8PB per storage system. With externally attached storage, Hitachi Virtual Storage Platform models VSP G1x00 and VSP F1500 can support up to 255PB of storage capacity. VSP G1x00 and VSP F1500 support 8Gb/sec and 16Gb/sec FICON (FICON Express8 and FICON Express8S) across all front-end ports for connectivity to the mainframe and 8Gb/sec Fibre Channel for connecting external storage. VSP G1x00 and VSP F1500 support high-performance FICON (z/hpf) for z/os. On the back end, VSP G1000 supports 6Gb/sec SAS protocol while VSP G1500 and VSP F1500 both support 12Gb/sec SAS.

17 WHITE PAPER 17 Hitachi Dynamic Provisioning Hitachi Dynamic Provisioning (HDP) for Mainframe optimizes performance through extremely wide striping and more effective use of storage through thin provisioning (see Figure 7). In other words, it allocates storage to an application without actually mapping the corresponding physical storage until it is used. This separation of allocation from physical mapping results in more effective use of physical storage with higher overall performance and rates of storage utilization. Dynamic Provisioning also enables Dynamic Volume Expansion (DVE) of 3390 volumes and FlashCopy SE for more efficient use of storage when creating local copies. Figure 7. Hitachi Dynamic Provisioning for Mainframe optimizes performance. Hitachi Dynamic Tiering Hitachi Dynamic Tiering (HDT) for Mainframe, an extension of HDP for Mainframe, enables the automatic movement of data between tiers. HDT for Mainframe provides an additional level of automated, optimized storage management by managing data across a full range of storage tiers from high-performance storage to low-cost storage. HDT for Mainframe moves highly accessed blocks of data to the highest tier storage and migrates less frequently accessed data to the lowest tiers according to simple policies. With HDT for Mainframe, there can be up to three storage tiers ranging from high-performance flash storage to low-cost storage, such as nearline SaS or virtualized external storage in the same storage pool. Tier creation is automatic, based on configuration policies, including media type and speed, RAID level and sustained I/O level requirements. This solution significantly reduces the time storage administrators have to spend analyzing storage usage and managing the movement of data to optimize performance and complements existing mainframe storage provisioning processes, such as DFSMS. Existing SMS storage groups and ACS routines can be aligned to differently tiered storage pools with pages of data being moved to the appropriate tier when needed rather than moving entire datasets.

18 WHITE PAPER 18 Hitachi is the first storage vendor to support virtualization of external storage behind an enterprise storage platform. HDT provides the ability for the system to automatically move blocks of data within data sets, to the most appropriate class of storage based on performance and access requirements. Hitachi Tiered Storage Manager for Mainframe Hitachi Tiered Storage Manager for Mainframe (HTSM) is a z/os software management product for Hitachi Dynamic Tiering for Mainframe that enables a user to control service levels based on performance and/or time to facilitate meeting mainframe service level agreements (SLAs). HDT policies managed through HTSM have the capability to hold selected data to a specified tier, regardless of the access patterns. The policies can also be defined to migrate selected data to a higher or lower tier on a predictable scheduled basis. HTSM enables Dynamic Tiering and Dynamic Tiering policies to be managed from the mainframe at the volume level or through DFSMS tools and constructs with ISPF or, optionally, REXX scripts. Hitachi Remote Replication Business continuity is more important than ever in today s business environment as demonstrated through the natural disasters and physical intrusion and destruction of IT resources over the last few years. A loss of business-critical data can force a company to its knees and even into bankruptcy. In addition, regulatory compliance requirements demand a business continuity and disaster recovery plan and infrastructure to support that plan or face stiff fines and business restrictions. Hitachi remote replication offerings provide the ability to copy critical data to off-site facilities either within a metropolitan area and/or to distant remote locations. The combination of the enterprise-level Hitachi Virtual Storage Platform models VSP G1x00 and VSP F1500 with Brocade s solutions to extend and optimize fabric connectivity facilitates the movement of your business-critical data over longer distances. Together, they enable and enhance your ability to support business continuity and disaster recovery. Hitachi TrueCopy Hitachi TrueCopy synchronous software provides a continuous, nondisruptive, host-independent remote data replication solution for disaster recovery or data migration over distances within the same metropolitan area. It provides a no-data-loss, rapid-restart solution (see Figure 8). For enterprise environments, TrueCopy synchronous software combined with Hitachi Universal Replicator on VSP G1x00 and VSP F1500 allows for advanced 3-data-center configurations. This combination includes consistency across up to 12 storage systems in each site for optimal data protection. Figure 8. Hitachi TrueCopy synchronous supports business continuity and disaster recovery efforts.

19 WHITE PAPER 19 TrueCopy synchronous supports business continuity and disaster recovery efforts, improving business resilience. It improves service levels by reducing planned and unplanned downtime of customer-facing applications. It enables frequent, nondisruptive disaster recovery testing with an online copy of current and accurate production data. TrueCopy synchronous can be seamlessly integrated into existing z/os environments and controlled with familiar PPRC commands or with Hitachi Business Continuity Manager software. Hitachi Universal Replicator Hitachi Universal Replicator provides asynchronous data replication across any distance for both internal VSP G1x00 and VSP F1500 storage and external storage managed by VSP G1x00 and VSP F1500 (see Figure 9). Universal Replicator provides enterprise-class performance associated with storage system-based replication. At the same time, it provides resilient business continuity without the need for remote host involvement, or redundant servers or replication appliances. Universal Replicator maintains the integrity of replicated copies without impacting processing, even when replication network outages occur or optimal bandwidth is not available. When compared to traditional methods of storagesystem-based replication, Universal Replicator leverages performance-optimized disk-based journals, resulting in significantly reduced cache utilization and increased bandwidth utilization. Universal Replicator ensures availability of up-to-date copies of data in up to three dispersed locations by leveraging the synchronous capabilities of Hitachi TrueCopy synchronous. In the event of a disaster at the primary data center, the delta resync feature of Universal Replicator enables fast failover and restart of the application without loss of data, whether at the local or remote data center. Figure 9. Hitachi Universal Replicator ensures availability of current copies of data in up to three dispersed locations.

20 WHITE PAPER 20 Universal Replicator can be integrated into an IBM GDPS environment, providing a much more cost-effective and complete recovery solution than the IBM alternative of z/os Global Mirror (XRC). With Universal Replicator and TrueCopy synchronous support of a 3-data-center replication solution, VSP G1x00 and VSP F1500 support delta resync, which is similar to but more efficient than z/os Metro or Global Mirror incremental resync. VSP G1x00 and VSP F1500 also support IBM z/os Basic HyperSwap, which is enabled by IBM Tivoli Productivity Center for Replication for System z Basic Edition (TPC-R). TPC-R enables the administrator to develop a z/os Basic HyperSwap configuration using VSP G1x00 and VSP F1500. Using VSP G1x00 and VSP F1500, the organization can create a z/os Basic HyperSwap plan. Create a 2-data-center configuration with TrueCopy synchronous or a 3-data-center configuration with TrueCopy synchronous, Universal Replicator and Business Continuity Manager. Initially, VSP G1x00 and VSP F1500 will support up to 12x12 (2DC) and 12x12x12 (3DC) systems at each site 1. VSP G1x00 and VSP F1500 with Universal Replicator will support a 4-data-center configuration and allow you to have two long asynchronous data paths and two synchronous paths. This solution offers you the ability to create multiple copies of data in many locations and reduce the impact of data migration. Hitachi Business Continuity Manager Hitachi Business Continuity Manager enables centralized, enterprise-wide replication management for IBM z/os mainframe environments. Through a single, consistent interface based on the Time Sharing Option/Interactive System Productivity Facility (TSO/ISPF) it uses full-screen panels to automate Hitachi Universal Replicator, Hitachi TrueCopy synchronous (including multisite topologies) and in-system Hitachi ShadowImage replication software operations. This software feature automates complex disaster recovery and planned outage functions, resulting in reduced recovery times. It also enables advanced, 3-data-center disaster recovery configurations and extended consistency group capabilities. Business Continuity Manager provides built-in capabilities for monitoring and managing critical performance metrics and thresholds for proactive problem avoidance. It also delivers autodiscovery of enterprise-wide storage configuration and replication objects, eliminating tedious, error-prone data entry that can cause outages. Business Continuity Manager integrates with the Hitachi replication management framework, Hitachi Replication Manager software, for replication monitoring and continuous operations in mainframe (and open system) environments. Multiplatform Support Hitachi Virtual Storage Platform models VSP G1x00 and VSP F1500 can support multiple operating systems at the same time. Although many mainframe organizations have been reluctant to share their storage platforms with open systems servers, the need to share storage is becoming more important: Organizations are implementing Linux on System z. In addition, the introduction of IBM zenterprise BladeCenter Extension (zbx) for mainframe processors enables Microsoft Windows to operate as part of zenterprise servers, VSP G1x00 and VSP F1500 can be configured to facilitate the isolation of disparate types of data. VSP G1x00 and F1500 easily supports organizations implementing private as well as some public clouds on zenterprise servers using either Linux on System z or Windows on zbx. Additionally, the FICON and Fibre Channel ports are completely separate and help ensure that critical mainframe data cannot be accessed directly by open systems servers or clients. 1 Check with your HDS Representative for currently supported configurations.