Mainframe Solutions

Dave Lytle

Five Tips for improving the value of your mainframe and storage ecosystem.

by Dave Lytle on ‎08-13-2014 12:32 PM (1,055 Views)

Tip 1: Use Brocade Fibre Channel switching devices to connect your FICON and/or FCP I/O mainframe and storage together.

 

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, i.e., DASD and tape platforms, along with a FICON switching platform which can match the impressive capabilities of zEnterprise. EMC, HDS, HP and/or IBM storage arrays paired with a Gen 5 DCX 8510 Director, together offer our customers the highest performing, most reliable, most scalable, and most cost-effective and energy-efficient products in the storage and networking industry for both FICON and FCP I/O. Efficiency and effectiveness at a lower price is the hallmark of Gen 5 motherboard-based switching platforms for FCP I/O traffic.

 

There are a wide range of reasons why the full value of the System z I/O subsystem is only realized when used with switched-FICON. For instance:

  • 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, a failure of the cable, a failure of the entire storage host adapter card, or a failure of 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 non-failing 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 (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.

 

There are 5 key technical reasons for connecting storage control units using switched FICON:

  • Overcome the maximum 40 buffer credit limitation imposed on FICON 8Gbps channel ports.
  • Build scalable Fan-in, Fan-out architecture designs for maximizing resource utilization.Switched-FICON.png
  • Localize potential failures to improved system availability.
  • Enable flexible connectivity for continued growth.
  • Leveraging z/OS and FICON technologies such as Node_Port ID Virtualization (NPIV), FICON Dynamic Channel Management (DCM) and zEnterprise Discovery and Configuration (zDAC).

 

To obtain a whitepaper about deploying switched-FICON please click the link below:

 

Tip 2: With switched-FICON, the availability of your I/O is dependent upon the robustness, scalability, failover capability and on-line repair capability of the fabric’s switching device so you should deploy directors at the heart of your I/O fabrics:

 

A director-class switch on zEnterprise is designed primarily to fulfill the role of core switch in a single switch Fibre Channel (FC) fabric or in a core-edge FC I/O topology. Because they sit at the center of a fabric, director-class switches are built to be modular, more scalable, perform better and be more resilient than the fixed-port switches (motherboard-based) that sit at the edge of a fabric and connect directly to storage such as physical tape drives. That is why the majority of mainframe shops deploy director-based I/O fabrics rather than fixed switch-based I/O fabrics.

 

Directors versus Switches.pngCompared with fixed switches, director switches have a larger number of ports; have built-in redundancy with non-disruptive upgrades of everything from power supplies to firmware; better scalability characteristics and provide unique, intelligent features such as inter-chassis links (ICLs).

 

Mainframe practitioners have a history of creating I/O infrastructures that have "five nines" availability. For FICON channel connectivity to mainframe-attached storage, these same organizations have a requirement for a FICON director platform that offers the same levels of RAS as the mainframe itself. The Brocade Gen 5 DCX 8510 is the ideal FICON director for these RAS requirements.

 

Because of their high bandwidth and throughput capabilities, director-class switches are also well-equipped to connect backup networks to backup storage media, such as tape libraries or disk arrays.

 

Tip 3: Always install the FICON Management Server (FMS) license so that you enable Control Unit Port (CUP) on FICON switching devices:

 

The CUP port (x”FE”) provides an in-band management interface defined by IBM that defines the Channel Command Words (CCWs) that the FICON host can use for managing the switch. The protocol used is the IBM version of the ANSI FC-SB3 single-byte command code specification, which defines the protocol used for transporting CCWs to the switch, and for the switch to direct data and status back.

 

The CUP is an optional licensed feature.

 

When you install the FMS license on a FICON switching device, and enable CUP on that device, you provide yourself with a lot of valuable tools:

  • RMF reports on FICON fabric statistics such as Frame Pacing Delay (the only way to get this information into RMF).
  • CUP is used to enable switch related hardware errors to be reported to z/OS against a device (domain ID) number.
  • If the switch is not defined as an I/O device through the use of CUP, and/or that I/O device is not online to z/OS, then switch related errors cannot be surfaced up to z/OS so that it can take appropriate actions.
  • Systems Automation (SA z/OS IOOPs ) can monitor and manage fabric components.
  • Date and Time on the FICON switching device can be set by z/OS for log and event synchronization.
  • The Allow/Prohibit Addressing Matrix (PDCM) becomes available for use.
  • FICON Dynamic Channel Management (DCM) can be used by Workload Manager.
  • The Port De-Commissioning Feature becomes available.

 

Click on the link below to read more about how CUP benefits a mainframe enterprise:

 

Tip 4: Make use of both local (ASIC) switching and backplane (central) switching to meet your business goals, objectives and SLAs:

 

Brocade has offered local switching (LS)for a decade on its switching products because LS provides the fastest I/O performance and reduces the amount of traffic that passes through the internal switch routing mechanism (e.g. core routing blades on Gen 5 directors). Local switching allows neighboring ports, on the same Application Specific Integrated Circuit (ASIC) to communicate without consuming valuable internal core / slot bandwidth. Since a single ASIC is doing the frame routing between two ports, LS reduces the amount of traffic that passes through the core freeing up that bandwidth to be used by other, backplane switching connected ports. Local switching results in lower I/O response time since it always runs at full line rate which in turn significantly reduces switch latency and helps to ensure that the chassis can achieve full line rate on all ports regardless of how they are connected.

 

Backplane and Local Switching.pngBackplane switching connects ports in a different way than local switching provides. Rather than using the ASIC itself to route I/O frames between two ports that it controls, central switching is used to route frames from the ingress port on one ASIC to the egress port on a different ASIC through valuable backplane bandwidth in the core. On directors this backplane bandwidth in the core is provided through the use of core router (CR) blades.

 

Backplane switching still results in a low I/O response time, only about 2 microseconds per frame, but since a frame passes through more hardware when using central switching it takes a little longer than local switching does. Through the use of cut-through frame routing backplane switching still reduces switch latency which results in a very low and consistent I/O response time.

 

Users choose local switching or backplane switching through how they cable ingress and egress ports on a chassis. When the ingress and egress ports are both controlled by a single ASIC then local switching is always utilized, otherwise central switching will occur.

 

Click on the link below to read some FAQs about Gen 5 8510 including local and backplane switching:

 

Tip 5: Make use of the Best FREE, high-value, FC Tool You Will Ever Use – SAN Health!

 

SAN Health.jpgBrocade SAN Health is a free software utility designed to securely audit and analyze FC FICON and FCP environments.

 

Simply put, the Brocade SAN Health family provides you with a straightforward and effective way to stop doing whiteboard SAN management. It provides information about your FICON SAN, FCP SAN, McDATA devices, Brocade devices and Cisco devices.

 

SAN Health provides you with powerful tools that help you focus on optimizing your storage network rather than manually tracking its components. A wide variety of useful features make it easier for you to collect data, identify potential issues, and check your results over time. As a result, you can greatly increase your productivity while enhancing your I/O fabric(s) operations.

 

To provide a comprehensive report about your storage networking environment, the free SAN Health Diagnostics Capture utility utilizes a data capture application and a back-end report processing engine. After it captures switch diagnostic data, the utility automatically generates a Visio topology diagram and a detailed "snapshot" report. This report contains summary information as well as specific details about SAN fabrics, switches, and individual ports. Other useful items include alerts, historical performance graphs, and recommended best practices. Use this utility in your own environment to perform critical tasks such as:

  • Take inventory of devices, switches, firmware versions, and fabrics.
  • Capture and display historical performance data.
  • Check zoning and switch configurations against best practices, including your own IOCP.
  • Assess performance statistics and error conditions.
  • Produce detailed graphical reports and diagrams.

 

Because it provides a point-in-time snapshot of your storage network, SAN Health Diagnostics Capture can be invaluable to your change-tracking process. For instance, you can use it to track traffic pattern changes in weekly or monthly increments. And with a built-in scheduler, you can run it after primary business hours for added safety and convenience.

 

Brocade customers and Brocade Partners can download the SAN Health Diagnostics Capture utility and SAN Health Professional for free:

  • SAN Health for HDS, HP and IBM:

o   http://www.brocade.com/services-support/drivers-downloads/san-health-diagnostics/download_san_health...

  • SAN Health for EMC:

o   www.emc.com/partnerships/technology-alliances/brocade-promo/install-san-health-pro.zip

 

To obtain much more information about Brocade’s SAN Health utility please click the link below: