This manual describes the features of the DECram for OpenVMS device driver (MDDRIVER). It includes information on how to determine which files should be stored on the device, how to configure the device, and how to use the device driver.
Revision/Update Information This document supersedes the DECram for OpenVMS User's Manual, Version 3.0
Operating System and Version: OpenVMS Alpha Version 7.2-1H1 or higher
Software Version: DECram for OpenVMS Version 3.1
Compaq Computer Corporation
Houston, Texas
© 2002 Compaq Information Technologies Group, L.P.
Compaq, the Compaq logo, Alpha, OpenVMS, VAX, VMS, and the DIGITAL logo are trademarks of Compaq Information Technologies Group, L.P. in the U.S. and/or other countries.
All other product names mentioned herein may be the trademarks or registered trademarks of their respective companies.
Confidential computer software. Valid license from Compaq required for possession, use, or copying. Consistent with FAR 12.211 and 12.212, Commercial Computer Software, Computer Software Documentation, and Technical Data for Commercial Items are licensed to the U.S. Government under vendor's standard commercial license.
Compaq shall not be liable for technical or editorial errors or omissions contained herein. The information in this document is provided "as is" without warranty of any kind and is subject to change without notice. The warranties for Compaq products are set forth in the express limited warranty statements accompanying such products. Nothing herein should be construed as constituting an additional warranty.
ZK5698
The Compaq OpenVMS documentation set is available on CD-ROM.
| Contents | Index |
This manual describes the DECram for OpenVMS device driver (MDDRIVER). You will find information on how to determine which files should be stored on the device, how to configure the device, and how to use the device driver.
This manual is intended for system managers who want to take advantage of the increased performance that results from using the DECram driver. You should be familiar with managing the OpenVMS Alpha operating systems before reading this manual.
This manual consists of three chapters, as follows:
See the following documents for information that is relevant to configuring and programming DECram disks:
The following documents contain information that will assist you in identifying files that can be stored on a DECram disk:
For additional information on OpenVMS products and services, access the Compaq website at the following location:
http://www.openvms.compaq.com |
Compaq welcomes your comments on this manual. Please send comments to either of the following addresses
| Internet | openvmsdoc@compaq.com |
|
Compaq Computer Corporation
OSSG Documentation Group, ZKO3-4/U08 110 Spit Brook Rd. Nashua, NH 03062-2698 |
Use the following World Wide Web address to order additional documentation:
http://www.openvms.compaq.com/ |
The following conventions are used in this manual:
| Ctrl/ x | A sequence such as Ctrl/ x indicates that you must hold down the key labeled Ctrl while you press another key or a pointing device button. |
| PF1 x | A sequence such as PF1 x indicates that you must first press and release the key labeled PF1 and then press and release another key or a pointing device button. |
| [Return] |
In examples, a key name enclosed in a box indicates that you press a
key on the keyboard. (In text, a key name is not enclosed in a box.)
In the HTML version of this document, this convention appears as brackets, rather than a box. |
| ... |
A horizontal ellipsis in examples indicates one of the following
possibilities:
|
|
.
. . |
A vertical ellipsis indicates the omission of items from a code example or command format; the items are omitted because they are not important to the topic being discussed. |
| ( ) | In command format descriptions, parentheses indicate that you must enclose the options in parentheses if you choose more than one. |
| [ ] | In command format descriptions, brackets indicate optional elements. You can choose one, none, or all of the options. (Brackets are not optional, however, in the syntax of a directory name in an OpenVMS file specification or in the syntax of a substring specification in an assignment statement.) |
| [|] | In command format descriptions, vertical bars separating items inside brackets indicate that you choose one, none, or more than one of the options. |
| { } | In command format descriptions, braces indicate required elements; you must choose one of the options listed. |
| bold text | This text style represents the introduction of a new term or the name of an argument, an attribute, or a reason. |
| italic text | Italic text indicates important information, complete titles of manuals, or variables. Variables include information that varies in system output (Internal error number), in command lines (/PRODUCER= name), and in command parameters in text (where dd represents the predefined code for the device type). |
| UPPERCASE TEXT | Uppercase text indicates a command, the name of a routine, the name of a file, or the abbreviation for a system privilege. |
| Monospace text |
Monospace type indicates code examples and interactive screen displays.
In the C programming language, monospace type in text identifies the following elements: keywords, the names of independently compiled external functions and files, syntax summaries, and references to variables or identifiers introduced in an example. |
| - | A hyphen at the end of a command format description, command line, or code line indicates that the command or statement continues on the following line. |
| numbers | All numbers in text are assumed to be decimal unless otherwise noted. Nondecimal radixes---binary, octal, or hexadecimal---are explicitly indicated. |
This chapter describes the DECram disk, which is a disk device created
in the physical memory area of a system. The operating system can read
from and write to a DECram disk, using standard OpenVMS disk I/O
operations, at access times much greater than those for standard
hardware disks. DECram disks use the DECram for OpenVMS device driver
(MDDRIVER).
1.1 Characteristics of a DECram Disk
The DECram driver allows you to create a disk in available physical memory and to read and write to that disk using standard OpenVMS disk I/O operations. This provides high-speed access to read-only data such as libraries, fonts, and command files. Additionally, the DECram disk can be used to hold temporary or scratch files that an application may require.
Other characteristics include the following:
Because DECram disk data is resident in main memory, the data will be lost if the host system fails or is shut down. Therefore, Compaq recommends using the DECram disk to store small, frequently accessed files such as temporary (scratch) files, or read-only files such as commonly used image files that reside permanently on a conventional disk unless the DECram disk is shadowed to a physical disk with OpenVMS host-based volume shadowing. |
Table 1-1 compares the features of a DECram disk to conventional disk features.
| DECram Features | Conventional Disk Features |
|---|---|
| Type of Storage Device | |
| A virtual disk that operates like a physical disk device. | Disk device. |
| Device Setup | |
| Set up once each time the system is rebooted. | Set up at installation time. There is rarely, if ever, a need to set up the device again. |
| Models Available | |
| Only one model, although the size can vary. Upgrading is not a consideration. | Many models, each fixed in size. Upgrading is always an important consideration. Options include price/performance, size of the disk, density, and seek time. |
| Use as a Storage Device | |
| Typically holds small files. Expansion is limited by the amount of system memory; cannot be used for offline storage. | Holds files of all sizes. Expandable to several orders of magnitude more than the amount of data that can be stored on a DECram disk. Some models have removable packs for secure offline storage. |
| Write Operations | |
| Stored data can be lost if system fails or is shut down. Writable, permanent files should generally not be placed on the DECram disk. However, if writable files must be placed on a DECram disk, then strongly consider the use of volume shadowing and journaling for data preservation. | Suitable for writable, permanent file storage. |
| Data Access Performance | |
| Extremely fast, limited only by CPU power and memory bandwidth, which are electronic in nature. | Speed limited by mechanical considerations, that is, seek time and rotation speed. Performance is also limited by other factors, such as interconnect bandwidth and controller features. |
Both DECram disks and disk caches improve system performance by providing faster access to data. However, they differ in how they function and in how they are used by the system. Table 1-2 describes these differences.
| DECram Features | Disk Cache Features |
|---|---|
| Type of Storage Device | |
| Operates like a disk device. | Operates as memory. |
| Device Setup | |
| Requires preloading files and using logical names to access these files on the DECram disk. | A disk cache is easily configured. The cache is usually transparent to the user. |
| Use as a Storage Device | |
| Typically holds entire files. | Typically holds only portions of files. |
| Can hold as much as 4,294,967,296 blocks of information on one disk with Version 3.0 or higher. The maximum amount of information on one disk that Version 2.3 can hold is 524,280 blocks for a VAX system and 67,108,864 blocks on an Alpha system. It is possible to create logically contiguous devices greater than either of these limits by creating multiple DECram devices and binding them together. | Capable of holding entire databases. |
| Both data blocks and file system blocks can be held. Consequently, opening and closing files is faster. | Disk caches often migrate to holding data blocks, as opposed to file system blocks. |
| Type of Data | |
| Choice of files to be held on the disk; subject to size constraints. | Choice of files to be held on disk caches not usually allowed. |
| Write Operations | |
| Not intended for permanent data storage. No write-back of data to permanent storage media. | Write operations write back data to permanent storage media. |
| Generally more efficient than writing to a disk cache; scratch files are created relatively quickly. | Writing to a disk cache is slower than writing to a DECram disk. Creating scratch files is a relatively slow process. |
| Local DECram disk files cannot survive a system shutdown. | Data blocks survive a system shutdown. |
| Shared DECram disk files can survive a system shut down. | |
| Data Access Performance | |
| Always reliable. Files are under the user's control and readily accessible. | Not always predictable. Access can vary by disk size and system load. (However, access is always more efficient than on comparable systems without any disk cache.) |
| Recommendations for Use | |
| Used for small, frequently accessed files such as application scratch files, system images, libraries, and DCL procedures. |
Disk caches should be used for databases, general user work files, and
files that increase in size.
Some disk caches use a protocol to write back data where the software is informed that a write operation is complete before ensuring the data is actually stored on the disk. Although write-back by a disk cache increases write performance to a level comparable to that of a write to a DECram disk, data can be lost if the system shuts down before the modified data is written back to disk. Users should be aware that the write-back characteristics of a disk cache can result in lost data. |
DECram Version 3.0 and higher is designed specifically to take advantage of some of the advanced capabilities of the newest Alpha systems, such as clustering and Galaxy shared memory. For example, one of the major differences between the capabilities of Version 2.3 and Version 3.0 and higher is that Version 3.0 moves the virtual device addressing from the S1 address into the S2 address space. This change allows for the creation and addressing of devices larger than 2.0 gigabytes (GBs).
DECram Version 3.0 was engineered with many new features available on AlphaServer systems. However, because VAX/VMS systems cannot use all of the new features in DECram Version 3.0 and higher, VAX/VMS systems are supported only through DECram Version 2.3. DECram Version 3.0 and higher is fully compatible with DECram Version 2.3. There can be any combination of these two versions of DECram in an OpenVMS cluster.
Because the geometry of DECram Version 3.0 and higher is different from Version 2.3, a BACKUP/PHYSICAL restore from Version 2.3 to Version 3.0 or higher will not work. The DECram Version 3.0 (or higher) driver does not support OpenVMS Fast IO (that is, the IO$_PERFORM QIO function). |
Any Alpha-based system can easily be upgraded to DECram Version 3.0 or higher. With Version 3.0 and higher you can use the new DECram command interface or continue using the same familiar commands from SYSMAN for creating, initializing, and mounting DECram disks.
In DECram Version 3.0 and higher, a disk is configured using the DECRAM> user interface to create and format the Random Access Memory (RAM) disk. The INITIALIZE command is then used to write the OpenVMS cluster file system to the RAM disk.
The following sections describe how these utilities and commands are
used to configure a DECram disk. Where applicable, and for the sake of
clarity, reference is made to both versions for issuing the commands.
2.1 Preparing to Configure a DECram Disk
Before creating a DECram disk, the system manager must determine the size of the DECram disk to be created. Each disk block allocated to a DECram device uses 516 bytes of memory. See Section 2.2.2 for information on how to allocate space. In addition, the system manager should assess the need for multiple DECram units based on application and user demands.
The system manager must allocate main memory from the system for the DECram disk. Therefore, a substantial amount of additional system memory may be required, depending on the size of files to be placed on the RAM disk. The amount of memory dedicated to a DECram disk is determined by the amount of memory required for a particular application and the system resources required for the DECram disk. Section 2.1.1 describes the system resources required for DECram disks.
DECram Version 3.0 (and higher) disks are created and formatted at the DECRAM> prompt and are initialized using the DCL INITIALIZE (INIT) command. If you configure DECram Version 3.0 or higher on OpenVMS Alpha (Version 7.2-1H1 or higher), you can generate a DECram startup procedure to set up the disk and copy any required files to it. Usually, this procedure is called from the system startup procedure SYS$MANAGER:SYSTARTUP_VMS.COM.
Be sure to check for disk errors after issuing any DECRAM command. Not
all errors are returned to the user interface. Errors specific to a
device are sent to the system error log. Type SHOW DEVICE MD at the DCL
prompt to see if there were any device errors generated as a result of
DECram command. You will need to use an error log analyzer tool to
recover the error. However, the errors are logged in ASCII file format
so you can search for errors with an MD-E-FAILURE prefix in the
SYS$SYSROOT:[SYSERR]ERRLOG.SYS file.
2.1.1 Resource Requirements
The DECram for OpenVMS device driver is a set of routines and tables that the OpenVMS operating system uses to process an I/O request for a DECram disk device. In addition to the nonpaged pool requirements for the driver and the I/O database structures (see the OpenVMS VAX Device Support Manual available on the OpenVMS Documentation CD-ROM), the DECram disk uses system memory, taken from the free page list, as device storage. You can specify the amount of memory required during the configuration procedure.
On OpenVMS Alpha systems (Version 7.2-1H1 or higher) running DECram
Version 3.0 or higher, the only requirements are that a DECram disk
must have 516 bytes of free page list per block (512 bytes) of disk
space allocated.
2.1.2 SYSGEN Parameters
You may need to increase the SYSGEN parameter S2_SIZE by an amount
equal to the size of the DECram disk you are creating. For example, if
S2_SIZE=0 and you are creating a 1.0 GB DECram disk, increase S2_SIZE
to 1000.
2.2 Configuring DECram Disks
With DECram Version 3.0 or higher, you can configure a local DECram disk using either the command procedures for DECram Version 2.3 or the new command procedures for Version 3.0 using the DECRAM> interface.
To configure a DECram disk using Version 3.0 or higher, you must perform the following tasks in this order:
You can configure shadowed DECram disks on OpenVMS Cluster members running OpenVMS VAX Version 5.5 or higher, or OpenVMS Alpha Version 6.1 or higher.
The shadowing of a DECram disk to a real physical disk is strongly recommended when using OpenVMS Alpha Version 7.2-1H1 or higher. The reason for this recommendation is to avoid a loss of data during an incomplete write operation. DECram writes data one block at a time. If the system experiences a power loss or other interruption of service and the system is in the process of writing, the block write would not complete successfully and the data would be lost. There is no way to recover the block of information because DECram data is volatile. One way to protect against this condition is to shadow the DECram disk with a physical disk so that data can be saved to a non-volatile medium and be recovered.
See the Volume Shadowing for OpenVMS manual for complete information about creating and
using host-based shadow sets.
2.2.1 Creating Local and Shared DECram Disks
When you upgrade to DECram Version 3.0 or higher, you can create disks to be shared by several nodes in a cluster by using the DECRAM> command interface. Detailed information about creating local and shared disks using the DECRAM> CREATE command is in Section 2.2.2.2 and Section 2.2.2.2.1.
Alternatively, you can continue to create local disks using the SYSMAN utility that was supported in earlier versions of DECram:
|
SYSMAN> IO CONNECT ddcuuuu:/DRIVER=SYS$MDDRIVER/NOADAPTER |
where:
ddcuuuu: represents the device name.
dd
The device code is always MD for a disk device.c
The controller designator can be any letter, A through Z.uuuu
The unit number can be any value from 0 to 9999.
$ RUN SYS$SYSTEM:SYSMAN SYSMAN> IO CONNECT MDA0:/DRIVER=SYS$MDDRIVER/NOADAPTER SYSMAN> EXIT |
The command in this example creates device MDA0 and connects it to the DECram driver (MDDRIVER).
See online help or the OpenVMS System Management Utilities Reference Manual for more information about the
SYSMAN command IO CONNECT.
2.2.2 Allocating and Deallocating Space for a DECram Disk
With DECram Version 3.0 and higher, there is no longer a limit on system space per block of disk space allocated.
A Galaxy configuration is required for creating a DECram shared disk. A stand-alone system in a Galaxy can access the shared disk but a Galaxy and an OpenVMS cluster is required to access a shared disk from multiple instances in the Galaxy or multiple nodes in the cluster. Therefore, shared disks are supported only on OpenVMS Alpha Version 7.2-1H1 (or higher) on systems that are capable of supporting Galaxy shared memory. For more information about Galaxy software, see the OpenVMS Alpha Partitioning and Galaxy Guide.
For more information on local or shared disks for DECram Version 3.0
(or higher), see Section 2.2.2.2.
2.2.2.1 Allocating and Deallocating Space Using INITIALIZE
You can allocate or deallocate memory space for a local DECram disk by using the DCL command INITIALIZE.
Any DECram disk that is mounted cluster-wide must be dismounted from every node before it can be initialized. |
|
INITIALIZE ddcuuuu: volume-label /SIZE=n |
where:
ddcuuuu:
The device name used in the SYSGEN command CONNECT to create the device (see Section 2.2.1).volume-label
A unique disk label up to 12 characters maximum./SIZE=n
The amount of memory (in blocks or 512 byte chunks) to be allocated to this device.
Restriction
With DECram for OpenVMS Version 3.0 or higher, the limit on the DECram disk size has been extended to 4,294,967,296 blocks on OpenVMS Alpha systems running Version 7.2-1H1 or higher. Under DECram Version 2.3, the maximum disk size on an OpenVMS Alpha system is 67,108,864 blocks and on an OpenVMS VAX system, disk size is limited to 524,280 blocks. It is possible to create logically contiguous devices geater than either of these limits by creating multiple DECram devices and binding them together.To deallocate memory used by a DECram disk, specify /SIZE=0 in a subsequent INITIALIZE command. This returns all memory resources to the system.
Warning
If you initialize a DECram disk, all data on the disk is lost.
Refer to the online help available through DECram or see the OpenVMS DCL Dictionary for more information on the INITIALIZE command.
$ INITIALIZE/INDEX=BEGINNING MDA0 FASTRAMDISK /SIZE=300 |
The command in this example initializes the device MDA0 with a size of 300 blocks. The /INDEX=BEGINNING qualifier places the index file for the volume's directory structure at the beginning of the volume. The volume label is FASTRAMDISK. The following command deallocates the same disk by setting /SIZE to 0.
$ INITIALIZE MDA0 FASTRAMDISK /SIZE=0 |
The following error messages can be returned for the DCL command INITIALIZE:
Initialization Failure and Recovery
If disk initialization fails due to insufficient resources, you can attempt recovery by performing one or more of the following steps:
With DECram Version 3.0 and higher, you can allocate space for a local DECram disk or, on OpenVMS Alpha Version 7.2-1H1 (or higher), for a disk in Galaxy shared memory. Local memory is accessible only from the local node or instance. Shared memory is accessible from any Galaxy instance that attaches to the shared memory section. (See Section 2.2.2.2.1 for more information about allocating and deallocating shared disks.) For details about Galaxy, refer to the OpenVMS Alpha Partitioning and Galaxy Guide.
To allocate space, you must have the following privileges: AUDIT, CMKRNL, SYSLCK, SYSPRV, and PHY_IO.
Enter the following command from the DCL prompt:
|
DECRAM CREATE DISK device-name /CAPACITY=blocks [/qualifier] |
where:
device-name
The name of the virtual memory disk that you want to create or modify in local memory or in Galaxy shared memory.The disk name takes the form ddcuuuu, where dd is always MD, the controller c can be a letter from A through Z, and the disk unit number uuuu can be any number in the range 0 through 9999 (for example, MDC256).
/CAPACITY=blocks
This required qualifier specifies the size in blocks (512 bytes=1 block) of the virtual disk to be created or modified. The capacity block limit in DECram Version 3.0 and higher is now 8,388,608 blocks./[NO]ALLOCLASS=class-number
This optional qualifier specifies an allocation class to be permanently associated with the controller. This qualifier is allowed for SHARED memory type disks only. The default allocation class value is 99.
Note
Once you use /ALLOCLASS to associate a controller letter with a specific allocation class, this association remains in effect as long as the shared memory section and the MDRECOVER.DAT file both exist./[NO]PERSIST
This optional qualifier controls whether or not the disk is restored after a reboot. The default qualifier, if not stated, is /PERSIST.Following are the three ways to restore a disk:
- Execute the SYSMAN command MDRECOVER image at startup time (see Section 2.3.1).
- Execute the DCL command DECRAM>RECOVER (see Section 2.3.2).
- Execute the DECRAM>CREATE DISK command again (see Section 2.2).
/[NO]SERVE
This optional qualifier controls whether or not disk serving is enabled for the OpenVMS Cluster system. The default qualifier, if not stated, is to /NOSERVE.Once a disk has been specified as /SERVE, it cannot be selected as /NOSERVE. If you create disks on one node with the /NOSERVE qualifier and then try to mount that disk cluster wide with the MOUNT/CLUSTER command then the process where the MOUNT/CLUSTER command was issued will hang. A DECram disk can be changed from NOSERVE to SERVE status at any time by issuing the original DECram CREATE DISK command with the SERVE qualifier. If you intend to do a cluster-wide mount then you must create the DECram disk using the /SERVE qualifier.
A DECram disk can be changed to be served to the cluster at any time by issuing the same DECRAM CREATE DISK command with the /SERVE qualifier. It is not necessary to delete the DECram disk first. This is the only attribute that can be modified while the disk size is greater than zero.
Shared memory is accessible from any Galaxy instance that attaches to the shared memory section. To create a shared disk, you must specify the /MEMORY=SHARED qualifier in the CREATE command. (The CREATE command is described in Section 2.2.2.2.) If you do not specify /MEMORY, it defaults to the equivalent of /MEMORY=LOCAL. You must determine the type of memory to create based upon the type of data structures needed to support the applications and users on the system. Once you create a specific memory type, there is overhead memory that is not deallocated when the DECram disk size is set to zero.
Shared disks may need to be initialized only once---the first time they are created. For example, after you execute a CREATE command for the disk, you would then execute a command such as the following:
|
When a DECram shared disk is created on one instance it automatically appears on all other instances in the Galaxy if at least one shared disk has been created on each of the instances in the Galaxy. For example, if DECram was installed on each instance in the Galaxy with the shared disk option during installation then one shared disk would have been created on each node.
You cannot change the size of a shared disk while other nodes are attached to it. You must set the size of the shared disk to zero on all nodes before you can change the size of a shared disk.
DECram recovers shared disks after a power failure and attaches them to existing shared memory regions. This instantly restores DECram data that was available prior to the crash. The recovery data is located in a file called MDRECOVER.DAT and works best if this file is located on a common VMS system disk. If each node has its own MDRECOVER.DAT file then it might not restore the same disks as other Galaxy instances. If a common system disk is not possible then it is up to the system manager to keep the multiple copies of the MDRECOVER.DAT file consistent by copying the master to all instances at SYS$COMMONH:[SYSMGR]MDRECOVER.DAT.
$ DECRAM CREATE DISK MDB150 /CAPACITY=600 /MEMORY=SHARED /ALLOCLASS=5 |
The command in this example creates the shared disk MDB150 with a size of 600 blocks. The shared disk controller is permanently associated with allocation class 5.
The following example deallocates the same disk by setting the /CAPACITY qualifier to 0.
$ DECram CREATE DISK MDB150 /CAPACITY=0 /MEMORY=SHARED /ALLOCLASS=5 |
When deallocating, you must specify the command line qualifiers in exactly the same way shown in the example above when creating the DECram disk except with the qualifier /CAPACITY=0. If you do not do this, you will get an unsupported error message and the disk will NOT be deallocated. |
In general an unsupported error message, %SYSTEM-E-UNSUPPORTED,
unsupported operation or function, means that you have entered an
invalid command and/or command qualifier. You should review commands
and qualifiers in DECram help and then re-enter the new command and/or
qualifier.
2.2.3 Mounting a DECram Disk
The method for mounting a DECram disk varies, depending on whether the disk is local or shared, as follows:
By default, VAXcluster I/O Cache and Extended File Cache (XFC) both disable caching on DECram disks, so you do not need to specify /NOCACHE with the MOUNT command. Specifying MOUNT/CACHE can improve performance on MSCP served DECram disks, but VAXcluster I/O Cache and XFC ignore /CACHE for DECram disks. |
You can mount a local DECram disk using the following syntax of the DCL command MOUNT:
|
MOUNT device-name: volume-label |
device-name
The device name used to create the device in the SYSMAN command IO CONNECT on Alpha systems or at the DECRAM> prompt using the CREATE command.volume-label
The unique disk label used in the INITIALIZE command.
$ MOUNT MDA0: FASTRAMDISK |
The command in this example mounts the device MDA0. The volume label is FASTRAMDISK.
Data on the DECram disk is preserved if you dismount and then mount the disk. |
See the OpenVMS System Management Utilities Reference Manual for more information on the MOUNT command.
2.2.5 Mounting a Shared DECram Disk
Use the appropriate syntax of the DCL command MOUNT to mount a shared disk for single or clustered systems:
For a single standalone system:
|
MOUNT/SYSTEM MDB150 FASTRAMDISK |
For a clustered system:
|
MOUNT/CLUSTER MDB150 FASTRAMDISK |
The DECram disk then can be accessed by any instance that is attached to the shared disk region.
$ MOUNT MDB150 VOLUME-LABEL |
The command in this example mounts the shared device MDB150. If you want to share the RAM disk, then you must specify the /SYSTEM or /CLUSTER qualifiers.
Data on the shared DECram disk is preserved if you dismount and then mount the disk. In addition, data is also preserved on the DECram disk even if the Galaxy instance crashes, as long as another instance is attached to the shared region. |
See the OpenVMS System Management Utilities Reference Manual for more information on the MOUNT command.
2.3 Restoring an OpenVMS Alpha Version 7.2-1H1 (or Higher) DECram Disk
Restoring a disk refers to the process of recovering a disk before it can be used. Once a disk is restored, you must still initialize the disk before it can be used.
There are three ways to restore all RAM disks:
The MDRECOVER.EXE image restores memory-based disks after a power failure or system crash. MDRECOVER.EXE uses memory disk data in SYS$COMMON:[SYSMGR]MDRECOVER.DAT to recover memory disks with the /PERSIST qualifier set.
Separate MDRECOVER.DAT files may not have the exact same list of disks in them and that a recovery from one node may not restore the disks of another cluster node. A common MDRECOVER.DAT file is required for consistent cluster-wide memory disk recovery. This file is invoked automatically when the system is rebooted. |
Issuing the "PRODUCT INSTALL DECRAM" statement at the DCL prompt invokes the POLYCENTER Software Installation procedure. POLYCENTER then installs the following command in the MDRECOVER.DAT file for execution by OpenVMS at system startup time:
|
MCR SYSMAN STARTUP ADD FILE MDRECOVER.EXE/PHASE=LPMAIN/MODE=DIRECT |
This command is issued automatically during installation and results in
the restoration of the RAM disk(s) automatically after a power failure,
system maintenance, or interruption of service for any reason.
2.3.2 Restoring a Disk Using the DCL command DECRAM RECOVER
The following command can be issued to recover a DECram disk:
|
DECRAM RECOVER |
The DECram command RECOVER reads from the MDRECOVER.DAT file located in the SYS$COMMON:[SYSMGR] area and restores each disk in the file that has the /PERSIST qualifier set.
The following DECram command recovers all the disks with /PERSIST set:
|
DECRAM RECOVER |
2.4 Determining Allocation of DECram Device Resources
The SHOW DEVICE command is used to show the resources that are allocated for existing RAM disks.
The DECRAM SHOW DISK command shows all of the disk resources that will be allocated on the next disk recovery when using the SYS$COMMON:[SYSMGR]MDRECOVER.DAT file. That command includes all disks with or without the qualifier /PERSIST being set.
If MDRECOVER.DAT is a cluster wide common file, it will not list local disks of other cluster nodes with different allocation classes. |
When using DECram Version 3.0 or higher, you can display information about a virtual disk from both the DECRAM$RECOVER.DAT and MDRECOVER.DAT recovery files by issuing the DCL command DECRAM SHOW DISK.
|
DECRAM SHOW DISK [device-name] |
device-name
The device name used in the CREATE DISK command (see Section 2.2.2.2). If no device name is specified, a brief status is displayed for all disks in the recovery files.
When the command includes a disk specification, the following information is displayed about the disk:
When no device name is specified, only the disk name, size, and volume label (if available) is displayed for all disks in the recovery files.
If no output is displayed, no DECRAM$RECOVER.DAT or MDRECOVER.DAT files exist.
Reference the online help information available through the OpenVMS DCL HELP command or see the OpenVMS DCL Dictionary for more information on the SHOW DISK command.
$ DECRAM SHOW DISK |
This command does not specify a device, so a brief summary of information is displayed for all devices in the DECRAM$RECOVER.DAT and MDRECOVER.DAT recovery files.
From SYS$COMMON:[SYSMGR]MDRECOVER.DAT
Disk $4$MDA0 Size 110 Label MDA0
Disk $4$MDF0 Size 100 Label MDF0MDF0
Disk $4$MDF1 Size 100 Label MDF1MDF1
Disk $4$MDF2 Size 100 Label MDF2MDF2
Disk $99$MDE0 Size 100 Label MDE0
|
$ SHOW DISK MDA0
From SYS$COMMON:[SYSMGR]MDRECOVER.DAT
Disk $4$MDA0 Size 110 Label MDA0
Local memory, persist, serve, writebuffered.
|
This command displays full information about device MDA0.
Never delete the MDRECOVER.DAT file. If you delete the file, then you will not be able to recover any disks after a power failure. |
The algorithm used by shadowing to select shadow set members for an IO operation is based primarily on member queue lengths. Although the queue lengths of all shadow set members will almost always be the same, DECram is expected to do more IO since it returns reads and writes much more quickly than a physical disk. One way to ensure that the application reads only from shadow set DECram disks is to do IO local to the DECram disk while the other shadow set members are served.
If you use the INIT/SHADOW DCL command to initialize a DECram disk followed by a MOUNT/SHADOW command, be aware that OpenVMS volume shadowing will not execute a full copy operation. To execute a full copy operation of shadowed disks and make the data fully consistent between shadow set members, you should issue an INIT/ERASE command after the INIT/SHADOW command. |
To mount a shadowed DECram disk, use the appropriate format with the DCL command MOUNT for single or clustered systems:
|
MOUNT DScuuuu/shadow=($n$device-name:) volume-label |
For a single standalone system:
|
MOUNT/SYSTEM DSA0: /SHADOW=($3$MDB100, $3$MDA0) FASTRAMDISK |
For a clustered system:
|
MOUNT/CLUSTER DSA0: /SHADOW=($3$MDB100, $2$MDA0) FASTRAMDISK |
where:
DS
Represents the controller code identifier, which is always DS for a shadowed disk.c
Represents any single letter, A to Z.uuuu
Represents any numeric value between 0 and 9999.n
Represents the allocation class for this OpenVMS system.device-name
The full device-name returned by SHOW DEVICE device-name (see Section 2.2.1).volume-label
The unique disk label used in the INITIALIZE command.
$ MOUNT DSA0 /SHADOW=($2$MDA1000:) FASTRAMDISK |
The command in this example mounts the host-based shadow set DSA0 with the single member $2$MDA1000.
The volume label is FASTRAMDISK.
Data on the shadowed DECram disk is preserved if you dismount and then mount the host-based shadow set. |
See the manual Volume Shadowing for OpenVMS for more information on allocation class
and the MOUNT command. OpenVMS System Management Utilities Reference Manual: M--Z contains additional information on
the MOUNT command.
2.5.2 Mounting a Shadowed DECram Disk--DECram Disk to Physical Disk or Partition
DECram can be used to shadow real physical devices or partitions provided there is enough physical memory available. You would do this in any situation where, in addition to data integrity, application speed and performance are also important.
For reads, the shadowing code will read from the DECram disk. Writes will be written to all devices and, therefore, will be much slower than a write to a DECram disk only. On power failure, the data will be safe because it is stored on physical media.
Volume Shadowing for OpenVMS does not currently freeze writes to the shadow set when the physical disk is no longer available. Therefore, be aware that with Version 3.0 or higher, if the physical disk fails, you will be writing information to a volatile disk and data may be permanently lost if the write is interrupted due to a power failure. You can reduce, but not eliminate, the risk of losing write data when one physical disk fails by adding two physical disks to the shadow set. |
If you have an existing shadow set that you know has more reads than
writes, placing a DECram disk in the shadow set can dramatically
improve overall performance. The reason for this performance
improvement is that a read operation does not require accessing a
physical disk because the information is available on the RAM disk.
Therefore, the time required to access the data is minimal. A write
operation in a shadow set comprised of a DECram disk still requires the
time needed to write the data out to the physical disk.
2.5.2.1 Creating the DECram Disk
Use the following command to determine the total number of disk blocks your physical device (or existing DSA device) has; this is the size you must use to create your DECram disk:
|
SHOW DEVICE/FULL $n$device-name: |
Use the DECRAM CREATE command to create your DECram disk with the /CAPACITY qualifier equal to total blocks. Then use the MOUNT command to add the DECram disk to the shadow set and initiate the shadow copy to the DECram disk. These commands are shown in the following example.
$ show dev dsa640/full
Disk DSA640:,device type MSCP served SCSI disk, is online, mounted, file-
oriented device, shareable, available to cluster, error logging is enabled.
Error count 0 Operations completed 8107
Owner process "" Owner UIC [SYSTEM]
Owner process ID 00000000 Dev Prot S:RWPL,O:RWPL,G:R,W
Reference count 1 Default buffer size 512
** Total blocks 1027362 Sectors per track 85
Total cylinders 756 Tracks per cylinder 16
Volume label "TST640" Relative volume number 0
Cluster size 3 Transaction count 1
Free blocks 1027203 Maximum files allowed 128420
Extend quantity 5 Mount count 2
Mount status System Cache name "_$84$DKC200:XQPCACHE"
Extent cache size 64 Maximum blocks in extent cache 102720
File ID cache size 64 Blocks currently in extent cache 0
Quota cache size 0 Maximum buffers in FCP cache 4610
Volume owner UIC [SYSTEM] Vol Prot S:RWCD,O:RWCD,G:RWCD,W:RWCD
Volume Status: subject to mount verification, file high-water marking,
write-back caching enabled.
Volume is also mounted on CSG6.
Disk $4$DUA640:, device type MSCP served SCSI disk, is online, member of shadow
set DSA640:, error logging is enabled.
Error count 0 Shadow member operation count 8126
Host name "HSJ50R" Host type, avail HSJ5, yes
Alternate host name "HSJ50G" Alt. type, avail HSJ5, yes
Allocation class 4
Disk $4$DUA642:, device type MSCP served SCSI disk, is online, member of shadow
set DSA640:, error logging is enabled.
Error count 0 Shadow member operation count 8134
Host name "HSJ50R" Host type, avail HSJ5, yes
Alternate host name "HSJ50G" Alt. type, avail HSJ5, yes
Allocation class 4
$ DECRAM CREATE DISK MDA1642/CAP=1027362/SERVE/ALLOCLASS=84
$ mount/cluster dsa640/shad=$84$MDA1642: TST640
%MOUNT-I-MOUNTED, TST640 mounted on _DSA640:
%MOUNT-I-SHDWMEMCOPY, _$84$MDA1642: (CSG84) added to the shadow set with a copy
operation
%MOUNT-I-ISAMBR, _$4$DUA642: (HSJ50R) is a member of the shadow set
%MOUNT-I-ISAMBR, _$4$DUA640: (HSJ50R) is a member of the shadow set
$ show dev dsa640
Device Device Error Volume Free Trans Mnt
Name Status Count Label Blocks Count Cnt
DSA640: Mounted 0 TST640 1027203 1 5
$4$DUA640: (HSJ50R) ShadowSetMember 0 (member of DSA640:)
$4$DUA642: (HSJ50R) ShadowSetMember 0 (member of DSA640:)
$84$MDA1642: (CSG84) ShadowCopying 0 (copy trgt DSA640: 1% copied)
$
|
To mount a shadowed DECram disk, use the following format with the MOUNT command, using the existing shadow set device name:
|
MOUNT DScuuuu/shadow=($n$device-name:) volume-label |
where:
DS
Represents the controller code identifier, which is always DS for a shadowed disk.c
Represents any single letter, A to Z.uuuu
Represents any numeric value between 0 and 9999.n
Represents the allocation class for this OpenVMS Cluster member.device-name
The full device-name returned by SHOW DEVICE device-name (see Section 2.2.1).volume-label
The unique disk label used in the INITIALIZE command.
$ MOUNT DSA0 /SHADOW=($2$MDA1000:,$2$DUA546) FASTRAMDISK |
The command in this example mounts the host-based shadow set DSA0 with the members $2$MDA1000 and $2$DUA546. The volume label is FASTRAMDISK. See Section 2.5.2.2 for more information on mounting a DECram disk.
See the manual Volume Shadowing for OpenVMS for more information on allocation class and the MOUNT command. OpenVMS System Management Utilities Reference Manual: M--Z contains additional information on the MOUNT command.
This chapter describes applications that may benefit the most from
using DECram disks, how to identify and locate files that can be stored
on a DECram disk, and how using a DECram device might adversely affect
system operations.
3.1 Recommended Applications for a DECram Disk
There are two general classes of applications that benefit substantially from using a DECram disk:
Applications that frequently use system images execute a DCL command file in response to a user command. By placing the images and their associated libraries as well as heavily accessed data files on a DECram disk, execution speed improves. While it is possible to install some of the shared images as resident and thereby improve performance, the DECram disk can be used for data access, command procedures, or any other files that you simply want to run faster.
If the system fails or is shut down, you must reinitialize and then remount the local DECram disk. The system startup file must invoke a command procedure that can perform these functions. Shared disks are automatically created. A system startup file can mount the shared DECram disk for immediate use of saved data.
If system files are installed on a DECram device, you cannot dismount that DECram device while system files are installed or open. Installed system files are open as long as they are mapped into a process. |
Modular applications often accept input files, perform operations on
the data, and write output files. These output files are then read by
another program that produces other files, and so on. The application
runs with greater efficiency if these intermediate input and output
files are stored on a DECram disk. If the system fails or is shut down,
it is necessary to restart the application from the beginning, or from
a checkpoint if the application was a long-running continuous process.
3.1.1 Identifying Commonly Used Images
You can use the LIST command provided by the OpenVMS Install Utility (INSTALL) to identify commonly used images that may benefit from being stored on a DECram device. INSTALL tracks the number of times an image was accessed since installation.
By executing the INSTALL LIST command periodically and examining the output, you can determine the most frequently used images, as shown in the following example:
$ INSTALL LIST/FULL |
This command produces output that is similar to what follows:
DISK$DUA0:<SYS0.SYSCOMMON.SYSEXE>.EXE
ANALIMDMP;1 Prv
Entry access count = 0
Privileges = CMKRNL CMEXEC
AUTHORIZE;1 Prv
Entry access count = 0
Privileges = CMKRNL
CDU;1 Open Hdr Prv
Entry access count = 0
Current / Maximum shared = 0 / 0
Privileges = CMEXEC
COPY;1 Open Hdr Shar
Entry access count = 12
Current / Maximum shared = 0 / 2
Global section count = 1
DCL;1 Open Hdr Shar Lnkbl
Entry access count = 21
Current / Maximum shared = 0 / 3
Global section count = 1
|
The output shows that only the COPY.EXE and DCL.EXE files, as indicated
by the entry access count, have been accessed since the last
installation. This indicates that they might be potential candidates
for storage on a DECram device.
3.1.2 Using DECram Disks with a Workstation
Diskless workstations that access conventional disks over Ethernet can
also use a DECram disk to improve performance. Because the most
commonly used images are stored on the DECram disk, page faults and
image activation occur locally, rather than over the Ethernet. While
products such as XFC (a dynamic cluster caching product) can adjust
over time to provide a high hit rate for frequently accessed images,
only DECram can provide you with a 100 percent hit rate.
3.1.3 Storing SYS$SCRATCH Files on a DECram Disk
You can significantly improve system performance by using a DECram disk to hold scratch files produced by OpenVMS images, shareable images, and layered products.
Table 3-1 lists some of the OpenVMS images, shareable images, and layered products that open temporary files on SYS$SCRATCH.
If the system on which the device physically exists either fails or is taken out of service, any journal or scratch files created by the images, shareable images, and layered products listed in Table 3-1 will be permanently lost. |
| SYS$LIBRARY:ADARTL.EXE | SYS$SYSTEM:DTM.EXE |
| SYS$LIBRARY:CMS$$MAILSHRP.EXE | SYS$SYSTEM:EDT.EXE |
| SYS$LIBRARY:CONVSHR.EXE | SYS$SYSTEM:MAIL.EXE |
| SYS$LIBRARY:EDTSHR.EXE | SYS$SYSTEM:NOTES.EXE |
| SYS$LIBRARY:MAILSHRP.EXE | SYS$SYSTEM:PCA.EXE |
| SYS$LIBRARY:NOTES$MAILSHRP.EXE | SYS$SYSTEM:SORTMERGE.EXE |
| SYS$LIBRARY:PPLSHR.EXE | SYS$SYSTEM:VERIFY.EXE |
| SYS$LIBRARY:SMGSHR.EXE | SYS$SYSTEM:VMSTAILOR.EXE |
| SYS$LIBRARY:SORTSHR.EXE |
To access SYS$SCRATCH files that are sent to a DECram disk, each user must have a directory on the DECram disk and define the logical name SYS$SCRATCH to point to that directory.
You can create a directory by using the following format at the DCL command CREATE/DIRECTORY:
|
$ CREATE/DIRECTORY MDA0:[directory] |
You must define the logical name SYS$SCRATCH to point to the directory on the DECram disk by using the following format:
|
$ DEFINE SYS$SCRATCH MDA0:[directory] |
where MDA0 represents the name of the DECram disk, and
directory represents the name of the user's directory
on the DECram disk.
3.1.4 Recovering Writable DECram Files
Compaq recommends that you store only small, frequently accessed files, such as temporary (scratch) files or read-only files, because data can be permanently lost if the node on which it exists fails or if the DECram device is reinitialized.
If you have stored and lost writable data, you can recover those files
by using OpenVMS RMS after-image journaling (AIJ). This is a layered
product and requires a usage license. It can be used to recover
critical, writable files stored on the DECram device. See the
RMS Journaling for OpenVMS Manual (available on the OpenVMS documentation CD-ROM) for more
information on RMS after-image journaling.
3.2 Using a Search List to Locate Files
You can use a search list to move a subset of files from a conventional disk to a DECram disk. A search list instructs the operating system to look for files on the DECram disk first, before looking for the files on a conventional disk. A search list is a logical name that has more than one translation, as shown in the following example:
$ DEFINE WORKFILE [WILSON.NOTES],[WILSON.WORK]
$ SHOW LOGICAL WORKFILE
"WORKFILE" = "[WILSON.NOTES]" (LNM$PROCESS_TABLE)
= "[WILSON.WORK]"
|
If you use the logical name CONSTRUCTION to point to the directory SYS$SYSDEVICE:[CONSTRUCTION], and you have moved only a subset of the files in that directory to the DECram disk, redefine the logical name CONSTRUCTION as a search list, as shown in the following example:
$ DEFINE/SYSTEM CONSTRUCTION "MDA0:[CONSTRUCTION]", - _$ "SYS$SYSDEVICE:[CONSTRUCTION]" |
Once the logical name has been redefined, any file lookup that uses the search list will examine each translation of the logical name to find the file. In this example, the CONSTRUCTION directory on the DECram disk will always be searched first. Defining a search list allows you to place a subset of the files addressed by a logical name on a DECram disk, while leaving the other files on the conventional disk. Any new file creations that use the search list will be created on the DECram disk because it is the first translation of the search list.
If you redefine system logical names and then install a product, the files that result from that installation could be stored on the DECram disk. However, these files will be lost if the system fails or is taken out of service. |
To prevent this loss, redefine the search list back to its former definition before installing the product. This deletes the DECram disk from the search list, as shown in the following example:
$ DEFINE/SYSTEM CONSTRUCTION "SYS$SYSDEVICE:[CONSTRUCTION]" $ SHOW LOGICAL CONSTRUCTION - _$ "CONSTRUCTION" = "SYS$SYSDEVICE:[CONSTRUCTION]" |
After the software installation completes, redefine the logical name to
add the DECram disk back to the search list.
3.3 Shadow Sets of DECram Devices
Compaq recommends that shadow sets of DECram devices consist of at least two members. A volume shadow set consisting of a single member that is a DECram device is subject to the failure scenario detailed in Section 3.4.3.
Starting with DECram Version 3.0 software, DECram devices can be shadowed to real physical disks or partitions. This is useful for applications that rely on data integrity and also require enhanced performance. The Volume Shadowing for OpenVMS software will read from the DECram device, rather than from the physical device, thereby providing a performance advantage.
The Volume Shadowing for OpenVMS software will write to both the
physical disk and DECram disk, so those writes will take longer than
reads. The advantage of Volume Shadowing for OpenVMS software writing
to both the physical disk and DECram disk is that data integrity is
preserved. If a node fails, all data on the physical disk or a
partition is still available and is not lost.
3.4 DECram Limitations
This section describes how using a DECram device can have a negative
effect on system operation.
3.4.1 Impact to Paging and Swapping
Using DECram disks reduces the size of the free page list. This may affect paging and swapping because of reduced available memory and could reduce or eliminate any performance gain in the I/O speed.
Some applications may require changing the values for the following system parameters for DECram disk memory requirements:
The following system parameters may need to be increased if you move installed images onto the DECram disk and reinstall them:
See the OpenVMS System Management Utilities Reference Manual: M--Z for information on system parameters.
3.4.2 I/O Delay Characteristics Affecting DECram Use
Using a DECram disk alters the execution characteristics of the system and of the applications that use DECram disks for I/O operations. I/O operations that use DECram disks are CPU intensive; the I/O is essentially a movement of data from system memory to user memory that executes in the system context. Therefore, applications that implicitly rely on I/O delay characteristics are not recommended for DECram disk use.
In some cases, applications that do not consume large quantities of CPU
resources because they are limited by I/O delay throughput will
suddenly become very CPU-bound. The applications become CPU-bound
because there is no waiting for I/O operations to complete. This change
in CPU consumption could then affect other users on the system.
3.4.3 Implications of Local Served and Shadowed DECram Devices
The failure characteristics of a DECram disk present an interesting problem in a served environment. The failure characteristics of a DECram disk are as follows:
When any DECram device has memory allocated to it (that is, it is initialized with a size of non-zero), a record of this is made in the SYS$COMMON:[SYSMGR]MDRECOVER.DAT file. Conversely, when the DECram device has memory released (that is, it is initialized to a size of zero), that memory deallocation is recorded in the same file.
During system initialization the SYS$COMMON:[SYSMGR]MDRECOVER.DAT file is interrogated. If any records of DECram devices that had memory allocated to them are found, the I/O database for those devices is restored.
When the MOUNT verification process on the other OpenVMS Cluster members interrogates these restored DECram devices, the MOUNT verification terminates and the applications currently using the devices receive an error status for any outstanding I/Os. No further use of these devices, such as dismounting them clusterwide and reinitializing them on the serving OpenVMS Cluster member, can be made.
| Index | Contents |