HP OpenVMS Systems

announcing the CIPCA adapter CIPCA support for Alpha Servers

Abstract:

This paper describes the CI-to-PCI adapter (CIPCA) that is supported in OpenVMS Alpha Versions 6.2-1H3 and 7.1 (the CIPCA is not supported in Version 7.0). CIPCA supports specific Alpha servers and OpenVMS Cluster configurations.

Table of Contents

» Product Overview

» Technical Specifications

» Configuration Requirements

» Figure 1 - CIPCA in a Mixed-Architecture OpenVMS Cluster

» Figure 2 - CIPCA in an Alpha OpenVMS Cluster

» Table 1 - CIPCA and CIXCD Performance

Product Overview

With the release of OpenVMS Alpha Version 6.2-1H2, Digital introduced a CI to PCI adapter (CIPCA) that was developed in partnership with CMD Technologies. With this adapter, Alpha servers that contain a combination of the PCI and the EISA buses can now connect to the CI.

CIPCA support for Alpha servers provides the following features and benefits to customers:

Lower entry cost and more configuration choices

If you require midrange compute power for your business needs, CIPCA enables you to integrate midrange Alpha servers into your existing CI OpenVMS Cluster.

High-end Alpha speed and power

If you require maximum compute power, you can use the CIPCA with both the AlphaServer 8200 systems and AlphaServer 8400 systems that have PCI and EISA I/O subsystems.

Cost-effective Alpha migration path

If you want to add Alpha servers to an existing CI VAXcluster, CIPCA provides a cost-effective way to start migrating to a mixed-architecture OpenVMS Cluster in the price/performance range that you need.

Advantages of the CI

The CIPCA connects to the CI, which offers the following advantages:

• High speed to accommodate larger processors and I/O-intensive applications.
• Efficient, direct access to large amounts of storage.
• Minimal CPU overhead for communication. CI adapters are intelligent interfaces that perform much of the I/O communication work in OpenVMS Clusters.
• High availability through redundant, independent data paths, because each CI adapter is connected with two pairs of CI cables.
• Multiple access paths to disks and tapes.

Figure 1 shows an example of a mixed-architecture CI OpenVMS Cluster that has two servers: an Alpha and a VAX.

Figure 1 - CIPCA in a Mixed-Architecture OpenVMS Cluster

As Figure 1 shows, you can use the CIPCA adapter to connect an Alpha server to a CI OpenVMS Cluster that contains a VAX server with a CIXCD (or CIBCA-B) adapter. This enables you to smoothly integrate an Alpha server into a cluster that previously comprised only high-end VAX systems.

Figure 2 shows another example of a configuration that uses the CIPCA to connect systems with the CI. In this example, each Alpha has two CIPCA adapters to increase performance. Also, the Alpha systems are connected to a high-speed FDDI interconnect, which provides scalable connectivity to PC clients and OpenVMS satellites.

Figure 2 - CIPCA in an Alpha OpenVMS Cluster

Figures 1 and 2 show that the CIPCA makes the performance, availability, and large storage access of the CI available to a wide variety of users. The CI has a high maximum throughput. Both the PCI-based CIPCA and the XMI-based CIXCD are highly intelligent microprocessor-controlled adapters that consume minimal CPU overhead.

Because the effective throughput of the CI bus is high, the CI interconnect is not likely to be a bottleneck. In large configurations like the one shown in Figure 2, multiple adapters and CI connections provide excellent availability and throughput. Although not shown in Figures 1 and 2, you can increase availablity by placing disks on a SCSI interconnect between a pair of HSJ controllers and connecting each HSJ to the CI.

Technical Specifications

The CIPCA is a two-slot adapter that requires one PCI slot and one EISA slot on the host computer. The EISA slot supplies only power (not bus signals) to the CIPCA. CIPCA's driver, PCAdriver, is supported in OpenVMS Alpha Versions 6.2-1H2, 6.2-1H3 and 7.1. Table 1 shows the performance of the CIPCA in relation to the CIXCD adapter.

Table 1 - CIPCA and CIXCD Performance

PERFORMANCE METRIC              CIPCA    CIXCD

Read request rate (I/Os)        4900     5500
Read data rate (MB/s)           10.6     10.5
Write request rate (I/Os)       4900     4500
Write data rate (MB/s)          9.8      5.8
Mixed request rate (I/Os)       4800     5400
Mixed data rate (MB/s)          10.8     9.2

Configuration Requirements

Following are the configuration requirements for the CIPCA:

• Configuration
  For OpenVMS Alpha Version 6.2-1H3, the CIPCA has the following configuration requirements:
  • Up to 16 CI host systems per star coupler.
  • Up to 3 CIPCA adapters per AlphaServer 2xxx system.
  • Up to 4 CPICA adapters per AlphaServer 4xxx system.
  • Up to 10 CIPCA adpters per AlphaServer 8xxx system with V6.2-1H3, and up to 26 CIPCA adapters per AlphaServer 8xxx system with V7.1.
  • AlphaServer 8400s can mix CIXCDs & CIPCAs in the same system.
  • Up to 32 CI products per star coupler (16 without CISCE).
  • Up to 96 OpenVMS Cluster member systems.
• Host systems
  The CIPCA can have the following hosts:
  • AlphaServer 8400
  • AlphaServer 8200
  • AlphaServer 4100
  • AlphaServer 4000
  • AlphaServer 2100
  • AlphaServer 2100A
  • AlphaServer 2000
• CI-connected hosts
  Any OpenVMS Alpha or VAX host using CIXCD or CIBCA-B.
• Storage controllers
  The CIPCA supports the following storage controllers:
  • HSJ30 and HSJ40 with HS0F Version 2.5 firmware or higher.
  • All HSCs except HSC50. (See Restrictions)
• Restrictions
  The following restrictions apply to CIPCA:
  • The HSC40 and HSC70 must have a Rev. F (or higher) L109 module. If your HSC40 or HSC70 does not meet this requirement, please contact your Digital Service representative.
  • The DIP switch settings on the CIPCA Link Module select cluster size and node address. The following instructions apply to setting the DIP switches for Link Module version A01 only:
    • If the cluster size is set to 16, do not set a CI adapter to node address 15 on that star coupler.
    • If the cluster size is set to 32, do not set a CI adapter to node address 31 on that star coupler. Also, one of the following must be done:

      Either do not set any CIPCA to node address 0, or do not set any CI adapter to node address 16.

    If these rules are not followed, arbitration timeout errors under heavy CI loads will cause CI path-bad errors and CI virtual circuit closures.

• Recommendation
  CIPCA uses a new, more optimal CI arbitration algorithm called Synchronous Arbitration instead of the older Asynchronous Arbitration algorithm. The two algorithms are completely compatible with each other. Under CI saturation conditions, both the old and new algorithms are equivalent and provide equitable round-robin access to all nodes. However, under less than saturation conditions, the new algorithm provides the following benefits:
  • Reduced packet transmission latency due to reduced average CI arbitration time.
  • Increased node-to-node throughput.
  • Complete elimination of CI collisions that waste bandwidth and increase latency in configurations containing only Synchronous Arbitration nodes.
  • Reduced CI collision rate in configurations with mixed Synchronous and Asynchronous Arbitration CI nodes. The reduction is roughly porportional to the fraction of CI packets being sent by the Synchronous Arbitration CI nodes.

  Support for Synchronous Arbitration is latent in the HSJ controller family. In configurations containing both CIPCAs and HSJ controllers, Digital recommends enabling the HSJs to use Synchronous Arbitration. The HSJ CLI command to do this is:

          CLI>  SET THIS CI_ARB = SYNC
  
  

  This command will take effect upon the next reboot of the HSJ.

For information about installing and operating the CIPCA, please refer to the hardware manual that came with your CIPCA adapter: CIPCA PCI-CI Adapter User's Guide (EK-CIPCA-UG).

» Back to the beginning and Table of Contents