 |
Claims  |
|
|
We claim:
1. A method for dynamically reorganizing placement of data in memory of a
computer system to improve retrievability of data stored therein
comprising the steps of:
monitoring the operation of said computer system memory;
detecting memory performance conflicts in said computer system memory;
identifying datasets in said computer system memory that must be relocated
to resolve said memory performance conflicts;
validating relocation of said identified datasets to resolve said memory
performance conflicts, comprising:
storing information describing a configuration of said computer system
memory,
storing a set of functional rules describing a data management function,
mathematically modeling at least a portion of said computer system memory,
and
collecting performance data on said identified datasets to validate
relocation of said identified datasets to resolve said memory performance
conflicts.
2. The method of claim 1 further comprising the step of:
generating said information describing said configuration of said computer
system memory.
3. The method of claim 1 wherein said step of detecting includes:
predicting potential performance conflicts in said computer system memory.
4. The method of claim 3 wherein said step of detecting further includes:
calculating statistical data from dataset read/write activity indicative of
the frequency of usage and locale of the datasets stored in said computer
system memory.
5. The method of claim 4, where said computer system memory includes a
plurality of DASD units, said step of identifying includes:
listing the ones of said DASD units that are in greatest and least conflict
with performance objectives.
6. The method of claim 5 wherein said step of identifying further includes:
selecting datasets from the ones of said DASD units listed as being in
greatest conflict and from the ones of said DASD units listed as being in
least conflict for exchange therebetween to balance the activity on these
listed DASD units.
7. The method of claim 6 further comprising the step of:
exchanging said selected datasets between said listed DASD units to reduce
the conflict on these listed DASD units.
8. The method of claim 4 where said computer system memory includes a
plurality of DASD units and a cache memory, said step of identifying
includes:
classifying all the datasets on a DASD unit that are good and bad
candidates for relocation to said cache memory.
9. The method of claim 8 wherein said step of identifying further includes:
listing the ones of said datasets, classified as good candidates for
relocation to said cache memory, that can be stored on one volume in said
DASD unit.
10. The method of claim 9 further comprising the step of:
transporting said listed datasets to said one volume in said DASD unit to
resolve said memory performance conflicts.
11. The method of claim 1 wherein said step of mode further includes:
generating data describing operational details of said configuration of
said computer system memory.
12. The method of claim 1 further comprising the step of:
transporting said identified datasets to alternate memory storage locations
to resolve said memory performance conflicts.
13. The method of claim 1 wherein said computer system memory further
includes at least one data channel connecting said computer system to a
control unit which manages a plurality of DASD units, said step of
detecting includes:
monitoring channel, control unit, device and dataset read/write activity in
said computer system memory.
14. The method of claim 1 wherein said step of modeling further includes:
storing a set of functional rules describing performance characteristics of
data storage devices that comprise said computer system memory; and
storing a set of functional rules describing desired service levels for
said data storage devices.
15. The method of claim 14 wherein said step of detecting includes:
predicting potential performance conflicts in said computer system memory;
and
identifying at least one of said data storage devices subject to said
predicted performance conflicts.
16. The method of claim 14 wherein said step of modeling further includes:
generating data describing operational details of said data storage devices
in said computer system memory.
17. A system for dynamically reorganizing placement of data in memory of a
computer system to improve retrievability of data stored therein
comprising:
means for monitoring the operation of said computer system memory;
means, responsive to said monitoring means, for detecting memory
performance conflicts in said computer system memory;
means, responsive to said detecting means, for identifying datasets in said
computer system memory that must be relocated to resolve said memory
performance conflicts;
means for validating relocation of said identified datasets to resolve said
memory performance conflicts, comprising:
means for storing information describing a configuration of said computer
system memory,
means for storing a set of functional rules describing a data management
function,
means for mathematically modeling at least a portion of said computer
system memory, and
means for collecting performance data on said identified datasets.
18. The system of claim 17 further comprising:
means for generating said information describing said configuration of said
computer system memory.
19. The system of claim 17 wherein said detecting means includes:
means for predicting potential performance conflicts in said computer
system memory.
20. The system of claim 19 wherein said detecting means further includes:
means for calculating statistical data from dataset read/write activity
indicative of the frequency of usage and locale of the datasets stored in
said computer system memory.
21. The system of claim 20, where said computer system memory includes a
plurality of DASD units, said identifying means includes:
means for listing the ones of said DASD units that are in greatest and
least conflict with performance objectives.
22. The system of claim 21 wherein said identifying means further includes:
means responsive to said listing means for selecting datasets from the ones
of said DASD units listed as being in greatest conflict and from the ones
of said DASD units listed as being in least conflict for exchange
therebetween to balance the activity on these listed DASD units.
23. The system of claim 22 further comprising:
means responsive to said identifying means for exchanging said selected
datasets between said listed DASD units to reduce the conflict on these
listed DASD units.
24. The system of claim 20 where said computer system memory includes a
plurality of DASD units and a cache memory, said identifying means
includes:
means for classifying all the datasets on a DASD unit that are good and bad
candidates for relocation to said cache memory.
25. The system of claim 24 wherein said identifying means further includes:
means for listing the ones of said datasets, classified as bad candidates
for relocation to said cache memory, that can be stored on one volume in
said DASD unit.
26. The system of claim 25 further comprising:
means responsive to said identifying means for transporting said listed
datasets to said one volume in said DASD unit to resolve said memory
performance conflicts.
27. The system of claim 17 wherein said modeling means further includes:
means for generating data describing operational details of said
configuration of said computer system memory.
28. The system of claim 17 further comprising:
means responsive to said identifying means for transporting said identified
datasets to alternate memory storage locations to resolve said memory
performance conflicts.
29. The system of claim 17 wherein said computer system memory further
includes at least one data channel connecting said computer system to a
control unit which manages a plurality of DASD units, said detecting means
includes:
means for monitoring channel, control unit, device and dataset read/write
activity in said computer system memory.
30. The system of claim 17 wherein said modeling means further includes:
means for storing a set of functional rules describing performance
characteristics of data storage devices that comprise said computer system
memory; and
means for storing a set of functional rules describing desired service
levels for said data storage devices.
31. The system of claim 30 wherein said detecting means includes:
means for predicting potential performance conflicts in said computer
system memory; and
means for identifying at least one of said data storage devices subject to
said predicted performance conflicts.
32. The system of claim 30 wherein said modeling means further includes:
means for generating data describing operational details of said data
storage devices in said computer system memory.
33. A system for dynamically reorganizing placement of data in memory of a
computer system to improve retrievability of data stored therein, wherein
said computer system memory is a hierarchial memory including a cache
memory and a plurality of DASD units, comprising:
means for monitoring the operation of said computer system memory;
means, responsive to said monitoring means, for detecting memory
performance conflicts in both said cache memory and said DASD units in
said computer system memory;
means, responsive to said detecting means, for identifying datasets in said
computer system memory that must be relocated to resolve said memory
performance conflicts;
means for validating relocation of said identified datasets to resolve said
memory performance conflicts, comprising:
means for storing information describing a configuration of said computer
system memory,
means for storing a set of functional rules describing a data management
function,
means for mathematically modeling at least a portion of said computer
system memory, and
means for collecting performance data on said identified datasets to
validate a relocation of said identified datasets to resolve said memory
performance conflicts.
34. The system of claim 33 further comprising:
means responsive to said identifying means for transporting said identified
datasets to alternate memory storage locations to resolve said memory
performance conflicts.
35. The system of claim 33 wherein said detecting means includes:
means for monitoring dataset read/write activity in said computer system
memory;
means for calculating statistical data from said monitored dataset
read/write activity indicative of the frequency of usage and locale of the
datasets stored in said computer system memory.
36. The system of claim 33 wherein said identifying means includes:
means for listing the ones of said DASD units that are most and least
utilized;
means responsive to said listing means for selecting datasets from the ones
of said DASD units listed as most utilized and from the ones of said DASD
units listed as least utilized for exchange therebetween to balance the
activity on these listed DASD units.
37. The system of claim 36 further comprising:
means responsive to said identifying means for exchanging said selected
datasets between said listed DASD units to balance the activity on these
listed DASD units.
38. The system of claim 33 wherein said identifying means includes:
means for classifying all the datasets on a DASD unit that are good and bad
candidates for relocation to said cache memory;
means for listing the ones of said datasets, classified as good candidates
for relocation to said cache memory, that can be stored on one volume in
said DASD unit.
39. The system of claim 38 further comprising:
means responsive to said identifying means for transporting said listed
datasets to said one volume in said DASD unit to resolve said memory
performance conflicts.
40. The system of claim 33 wherein said modeling means further includes:
means for storing a set of functional rules describing performance
characteristics of each of said DASD units; and
means for storing a set of functional rules describing desired service
levels for each of said DASD units.
41. The system of claim 40 wherein said detecting means includes:
means for predicting potential performance conflicts in said computer
system memory; and
means for identifying at least one of said DASD units subject to said
predicted performance conflicts.
42. The system of claim 40 wherein said modeling means further includes:
means for generating data describing operational details of said DASD
units.
43. A system for dynamically reorganizing the placement of data in memory
of a computer system to improve the retrievability of data stored therein,
wherein said computer system memory is a hierarchial memory including a
cache memory and a plurality of DASD units, comprising;
means for storing information describing the configuration of said computer
system memory;
means for storing a set of functional rules describing a data management
function;
means for mathematically modeling at least a portion of said computer
system memory;
means for detecting memory performance conflicts in said computer system
memory, including:
means for monitoring the dataset read/write activity in said computer
system memory;
means for calculating statistical data from said monitored dataset
read/write activity indicative of the frequency of usage and locale of the
datasets stored in aid computer system memory;
means, responsive to said detecting means, for identifying datasets in said
computer system memory that must be relocated to resolve said memory
performance conflicts, including:
means for listing the ones of said DASD units that are most and least
utilized;
means, responsive to said listing means, for selecting datasets form the
ones of said DASD units listed as most utilized and from the ones of said
DASD units listed as lest utilized for exchange therebetween to balance
the activity on these listed DASD units;
means for classifying all the datasets on a DASD unit that are good and bad
candidates for relocation to said cache memory;
means for listing the ones of said datasets, classified as good candidates
for relocation to said cache memory, that can be stored on one volume in
said DASD unit;
means, responsive to said identifying means, for exchanging said selected
datasets between said listed DASD units to balance the activity on these
listed DASD units;
means, responsive to said identifying means, for transporting said listed
datasets to said one volume in said DASD unit to resolve said memory
performance conflicts; and
means for writing said one volume from said DASD unit into said cache
memory.
44. A method of improving the data retrieval efficiency of a computer
system memory, wherein said computer system memory is a hierarchial memory
including a cache memory and a plurality of DASD units, comprising the
steps of:
recording information describing the configuration of said computer system
memory;
storing a set of functional rules describing a computer system memory
management function;
mathematically modeling at least a portion of said computer system memory;
monitoring the operation of both said cache memory and said DASD units;
detecting memory performance conflicts as a result of said monitoring;
identifying datasets in said computer system memory that must be relocated
to resolve said memory performance conflicts;
transporting said identified datasets to alternate memory storage locations
to resolve said identified memory performance conflicts. |
|
 |
|
 |
Claims |
|
|
|
Description |
|
|
FIELD OF THE INVENTION
This invention relates to computer system memory and, in particular, to an
intelligent storage manager for the data storage apparatus comprising the
computer memory system. This intelligent storage manager both monitors the
data retrieval efficiency of the computer system memory and either
modifies data set storage locations, or system storage configurations, in
response to memory performance conflicts detected in the computer system
memory to optimize the level of service provided to the user.
PROBLEM
It is a problem in data processing systems to efficiently manage the
storage of data sets on the computer system memory. Many large data
processing systems include a hierarchy of data storage devices that are
used by the computer systems contained therein. These data storage devices
range from off line magnetic tape cartridge data storage systems to on
line direct access storage devices which are directly connected to the
computer system, cache memory and extended store (a variation of
computer's main memory). Cache and extended store are fast on line memory
for use with data sets that are frequently read by the computer system. It
is difficult to dynamically allocate the location of the data sets in this
hierarchal arrangement of data storage devices so that the frequency of
data retrieval from these data sets matches the location and time wise
retrieval efficiency of the data storage device. It is advantageous for
computer system performance purposes to place the most frequently
retrieved data sets in cache memory and the most infrequently retrieved
data sets in the off line magnetic tape cartridge data storage systems.
From surveys done in the data processing field, in 1971 the median data
processing system installation had 1.2 gigabytes of direct access data
storage device capacity and these data storage devices had 450 data sets.
In 1980, these numbers were 23 gigabytes of direct access data storage
device capacity and 10,000 data sets. In 1988, surveys revealed that the
typical or median installation had 208 gigabytes of direct access data
storage device capacity which devices had 45,000 data sets stored therein.
There is a continuing significant growth in the capacity of direct access
data storage devices in the typical computer system. However, the
utilization of these direct access data storage devices has fallen from
80% in the late 1960s to approximately 45% in the present time frame.
Thus, there are significant increases in the cost of direct access data
storage device memory with a decreasing efficiency in the use of these
direct access data storage devices in the hierarchy of data storage
devices in the computer memory.
It is a commonly used metric in the computer system field that in order to
manage the computer system memory, it requires approximately 1 data
management employee to manage every 10 gigabytes of data in a computer
system installation. This data management person performs the functions of
I/O optimization, data set conflict avoidance, conflict resolution, data
set placement, and cache tuning in order to provide improved efficiency in
the use of the direct access data storage devices in the computer system.
Any improvement in the management of the data storage devices therefore
reduces the number of data management employees required to manage the
computer system memory. Newer data storage device architectures, e.g.
distributed databases, client server, etc., present new dimensions to the
storage performance management problem.
In addition to the problem of managing performance on a hierarchy of
storage devices locally attached to a computer system (or to a computer
complex consisting of multiple computer systems), network technologies and
emerging computer system architectures have added two additional
dimensions to the problem: geographical location and time. In a computer
network environment, data can reside on data storage devices located in
different geographic locations, e.g. Chicago and New York. Computer users
frequently have service requirements which may require data to be made
available on the local computer memory for a specific timeframe. A common
example is that of a regional sales office which needs access to their
client data for periodic update. Performance considerations may dictate
that this data be stored locally during update, e.g. in Chicago, and then
transferred to a central site, e.g. in New York City, for corporate
processing.
This increased complexity of the data processing system and all the
elements contained therein render the data storage device management task
beyond the capability of data management personnel. These tasks must be
automated and there presently exists no good solution to this problem.
SOLUTION
These problems are solved and a technical advance achieved in the art by
the intelligent storage manager for data storage apparatus. This apparatus
makes use of a knowledge based (expert) system to monitor in real time the
performance of the computer system data storage devices, identify memory
performance conflicts and resolve these conflicts by directing the
relocation of data sets, or user workloads, to other segments of the
computer system memory, and/or reconfiguring the computer system memory.
The subject apparatus includes a number of data bases which are used by the
expert system software to manage the computer system data storage devices.
One element provided in this apparatus is a set of data storage device
configuration data that provides a description of the various data storage
devices and their interconnection in the computer system. A second element
is a knowledge data base that includes a set of functional rules that
describe the data storage device management function. These rules indicate
the operational characteristics of the various data storage devices and
the steps that need to be taken to provide the various functions required
to improve the performance of the computer system memory. In addition to
these two data bases, this apparatus includes various mathematical models
to determine data relating to the operation of the data storage devices.
This additional data either cannot be directly measured or is not cost
effective to measure. The additional data is used to assist in the
identification of conflicts within the data storage apparatus. These
models are also used to predict the effect of various data management
actions and proposed solutions to identified conflicts. The models are
also used to manage the addition of data storage devices to the computer
system memory by identifying a plan to migrate existing data sets to these
additional data storage devices in a manner to optimize the service level
of the data processing system. Another element included in this
intelligent storage manager is an automatic system configuration
definition apparatus. This apparatus collects data generated within the
data processing system to identify the data storage devices and their
interconnection, without requiring user input. The determined
configuration data is used to populate the configuration database.
Operating on these two data bases, in an interactive manner with the
mathematical models, is an expert system, which is a computer program that
uses explicitly represented knowledge and computational inference
techniques to achieve a level of performance compatible to that of a human
expert in that application area or domain. The expert system includes an
inference engine, or technical equivalent, that executes the rules which
comprise the basic intelligence of the expert system. The inference engine
allows a virtually infinite number of rules or conditional statements to
be chained together in a variety of ways. The inference engine manages the
rules and the flow of data through those rules. Thus, the expert system
uses the data stored in the configuration data base and the knowledge data
base as well as data from monitoring of the actual performance of the data
storage devices in the computer system and data derived from model to
analyze, on a dynamic basis, the performance of the computer system data
storage devices.
The expert system identifies memory performance conflicts such as a
performance degradation of the computer system data storage devices due to
a plurality of computers in the computer system attempting to access a
common data storage device, or access to a given data set exceeding
acceptable service levels. The expert system identifies the performance
conflict as well as the data sets stored on these data storage devices
related to this conflict. Once the data sets and user workload involved in
the performance conflict are identified, the expert system determines
alternative memory storage locations for these data sets and activates
various software routines to transport these conflict data sets to the
alternative data storage locations. Alternatively the expert system
relocates the processing location of the user workload. The relocation of
these conflict data sets and/or user workload resolves the memory
performance conflict and improves the retrievability of the data stored on
these data storage devices. By performing the conflict identification and
resolution on a dynamic real time basis, the data storage devices of the
computer system are operated in a more efficient manner and the
retrievability of the data stored on these data storage devices is
significantly improved without the need for the data management personnel.
The intelligent storage manager continuously monitors and modifies the
performance of the data storage devices associated with the computer
system, even if the data storage devices are geographically dispersed. The
intelligent storage manager identifies and resolves conflicts in a manner
that accounts for the temporal, spatial and hierarchical characteristics
of the data processing system.
BRIEF DESCRIPTION OF THE DRAWING
FIG. 1 illustrates, in block diagram form, the architecture of the
intelligent storage manager;
FIG. 2 illustrates in block diagram form, the architecture of the cache
tuning and DASD performance optimizer elements of the intelligent storage
manager;
FIGS. 3 and 4 illustrate the elements that comprise the cache tuning and
DASD performance optimizer elements of the intelligent storage manager;
FIG. 5 illustrates an example of the objects used by the expert system
model apparatus.
DETAILED DESCRIPTION
The intelligent storage manager for data storage apparatus functions to
monitor the performance of a computer system memory and manage this
computer system memory in order to maximize performance. This function is
accomplished by the use of a knowledge based (expert) system that monitors
the activity of the various data storage devices in the computer system
memory, identifies memory performance conflicts and take steps to resolve
these conflicts in order to improve the performance of the computer system
memory. This apparatus includes a number of databases which are used by
the expert system software to manage the data storage devices.
DATA PROCESSING SYSTEM ARCHITECTURE
FIG. 1 illustrates a typical computer system in block diagram form, which
computer system includes the intelligent storage manager for data storage
apparatus. This computer system consists of a mainframe 101 which can be
any of a number of large computer systems such as an IBM 390 series, or an
Amdahl 5990 computer. Each of these mainframe computers can process on the
order of 10-20 MIPS per processor and are typically equipped with a
significant amount of disk storage memory, typically on the order of 200
gigabytes of data storage capacity. Frequently such systems are configured
into multi-system, multiple CPU complexes which share disk storage among
the individual systems, and/or data across networks. For the purpose of
illustration, assume that mainframe 101 is an IBM 3084 type of computer
running an operating system 102 that typically is the MVS/XA operating
system. These products are well known in the field and need no further
explanation herein. Mainframe 101 is interconnected via a plurality of
channel processors 110-119 to the data storage devices associated with
mainframe 101. For the purpose of illustration, a number of disk storage
devices 150-0 to 150-15 are illustrated in FIG. 1. Each of these disk
drives are a commercially available device such as the IBM 3390 class
device disk drive system or any other ones of the multitude of similar
devices. In a typical configuration, sixteen disk drives 150-0 to 150-15
are connected via associated data channels 140-0, 140-15 to a device
controller 130 which is typically an IBM 3990 class device type of
controller. Device controller 130 functions to multiplex the data being
read to and from the plurality of disk drives 150-0 to 150-15 onto a
channel path 120 which interconnects device controller 130 with channel
processor 110. Device controller 130 includes a cache memory 160,
including a non-volatile store 161 to provide greater data transfer
efficiency between mainframe 101 and the data storage devices 150-*, as is
well-known in disk technology.
The computer system can be equipped with a plurality of channel processors
illustrated in FIG. 1 as devices 110 to 119. Each channel processor 110 is
interconnected with mainframe 101 and provides mainframe 101 with access
to data storage devices connected to the channel processor. A multitude of
data storage devices can be connected to mainframe 101 in this manner. For
example, channel processor 119 is connected to a disk array 139 data
storage device. The other data storage devices include tape drives,
optical disk, solid state disk and other memory devices. While this block
diagram illustrates a general computer system configuration, it in no way
should be construed as limiting the applicability of the present
invention. In particular, various memory configurations can be used to
interconnect mainframe 101 with various data storage devices. In addition,
the data processing system can include a plurality of computer systems,
interconnected via network 191. The intelligent storage manager for data
storage apparatus 103 is included in mainframe 101 and operator terminal
170 is connected via data link 190 to mainframe 101.
MEMORY CONFIGURATION AND MANAGEMENT
From a data management standpoint, the computer memory illustrated in FIG.
1 is viewed as a data storage subsystem 180 that consists of sixteen
volumes of disk storage 150-0 to 150-15. Each volume of this data storage
subsystem contains a plurality of data sets or data files. These data sets
or data files are stored on one or more corresponding volumes in an
ordered fashion by dividing the volume up into sub-elements. For example,
in the 8380E disk drive, each volume consists of 1770 cylinders each of
which includes fifteen data tracks. Each data track can store 47K bytes of
data.
In operation, mainframe 101 accesses data stored on the data volumes in
data storage subsystem 180 by transmitting a data retrieval request via
the channel processor 110 associated with the data volume (e.g. - 150-0)
that contains the requested data. This data retrieval request is
transmitted by channel processor 110 over channel path 120 to device
controller 130, which in turn manages the retrieval of the data from the
designated data storage device 150-0. This data retrieval is implemented
by device controller 130 monitoring the rotation of disk drive 150-0 to
identify the beginning of the data set stored on a track of the disk drive
150-0. Once the disk has rotated sufficiently to place the beginning of
the requested data set under the read/write heads of the disk drive 150-0,
the data is transferred from disk drive 150-0 over data link 140-0 to
device controller 130. Device controller 130 temporarily stores the
retrieved data in cache memory 160 for transmission to mainframe 101 via
channel path 120 and channel processor 110.
It is obvious that this is an asynchronous data transfer in that mainframe
101 transmits data retrieval requests to various device controllers, where
these data retrieval requests are processed as the data sets become
available. If too many data retrieval requests are made via one route,
such as concentrating the requests on a specific volume or small subset of
volumes, the performance of the computer system memory significantly
degrades. Likewise if the inherent performance characteristics of the
device are inadequate for the service level required for a given data set
by a given user, a performance conflict exists. It is advantageous to
distribute the data retrieval requests throughout the various channel
processors, channel paths and disk drives to optimize the service time.
Thus, an intelligent distribution of the data retrieval requests among the
various volumes on each data storage subsystem and across all of the data
storage subsystems significantly improves the memory performance. There
are many choices available to improve memory performance such as moving
data sets from one volume to another or moving a volume from one data
storage subsystem to another, etc. It should also be noted that for
purposes of this patent disclosure, the terms "string" and "subsystem" can
be used interchangeably.
Another such improvement is the use of a cache memory, such as cache 160,
which enables the more frequently accessed volumes to load data in cache
memory 160 for more rapid data retrieval times. The volumes are selected
as being suitable for caching based on the normal access pattern of the
data sets that are stored therein. Any input and output accesses to the
data that use the cache memory 160 benefit from the cache memory 160 by
speeding up the data transfer time. In order to preserve data integrity,
any writes that are performed cause the data to be written directly to the
disk, or to a form of cache commonly called "non-volatile storage" 161 and
therefore these operations gain benefit from a cache memory 160 depending
on the specifics of each case. A good volume for caching is one that has a
high read rate, high activity and input/output accesses concentrated in a
small number of tracks rather than scattered across the entire disk.
Volumes that satisfy these conditions have a high hit ratio and therefore
the input/output reads often find the desired track in cache 160 and do
not have to spend the time retrieving the data set from the disk as
described above. Often, data storage subsystems contain no volumes that
are cachable or the activity of volumes that are cachable accounts for
only a fraction of the total input/output activity of the data storage
subsystem. Newer storage technologies, such as disk array, data "farms",
data servers, or single level storage, are represented by the disk drive
array 139. These technologies present a functional image of DASD volumes
to the channel 119. The physical storage of the data may vary considerably
from the traditional DASD storage architecture, frequently involving a
hierarchy of devices with varying performance characteristics. The
performance management of these types od devices is the same as managing
that of the mainframe 101. In fact the controllers for these types of
devices are typically powerful computer processors. The central controller
for such a device should be viewed as a mainframe 101 with attached
physical storage devices.
CACHE PERFORMANCE IMPROVEMENT
In order to gain full benefit of the cache, it is possible to reorganize
the data storage subsystem 180 in such a way as to increase the amount of
input/output accesses that go to specific cache volumes. A significant
problem is to reorganize the data storage subsystem 180 on a data set
level so as to separate the possible caching data sets from data sets that
would degrade the cache performance and, in the process, create entire
volumes that are suitable for storage in the cache memory 160. It is
obvious that restructuring the data set organization can entail a complete
restructuring of the volumes of the data storage subsystem 180 which
involves an enormous number of data set moves to accomplish. Instead of a
complete reorganization of the data storage subsystem 180, a better
approach is to identify places on the subsystem where conflicts exist
between possible cache data and data that is detrimental to the cache 160.
The conflict exists where there is good cache data and bad cache data on
the same volume: if the volume where cached in order to obtain the benefit
of caching the good data, the bad data would cause so much interference in
the cache 160 that the overall performance would actually deteriorate. The
intelligent storage manager locates conflicts and then moves the smallest
number of data sets in order to resolve the conflict. Using this approach,
a small number of moves provides a large increase in performance and
better utilization of the cache memory 160.
In order to determine which data sets are possible candidates for caching
and which data sets should not be cached, the input/output activity of the
data storage subsystem 180 is monitored over a period of time and data is
recorded on a data set basis. This stored data consists of the
input/output activity (the number of input and output operation to the
data set), the read percentage (percentage of operations that were reads),
and the locality of reference measure, which is defined as the number of
input and output to the data set divided by the number of unique tracks
covered by those inputs and outputs over the period monitored. Along with
these numbers, is stored the name of the data set, its size in bytes and a
value called the reason code that flags data sets that are not suitable
for moving. This data is used by the intelligent storage manager in
identifying the data sets that are to be moved and the target location for
these data sets.
INTELLIGENT STORAGE MANAGER FOR DATA STORAGE APPARATUS
The intelligent storage manager consists of a set of integrated data
storage management functions that use knowledge or expert system
techniques to monitor and automatically assist in the control of
performance of the direct access data storage sub-systems. FIG. 2
illustrates the principal components of the intelligent storage manager
103. Intelligent storage manager 103 interfaces with and runs on the MVS
operating system 102 of mainframe 101. In multiple system complexes, at
least the monitoring activity 171 must be running on all systems. Data
collected from all systems is processed on one of the systems by the
complete cache volume creation 105 and DASD performance optimizer 106. The
MVS operating system 102 provides device management, file management, test
management, processor management, communication management and a number of
other administrative functions to the intelligent storage manager 103.
In order to clarify terminology, the various aspects of knowledge systems
are discussed. Knowledge systems are computer based systems that emulate
human reasoning by using an inference engine or technical equivalent to
interpret the encoded knowledge of human experts that is stored in a
knowledge base. If the domain of the knowledge base is sufficiently narrow
and a sufficiently large body of knowledge is properly coded in the
knowledge base, then the knowledge system can achieve performance that
matches or exceeds the ability of a human expert. In such a case, the
knowledge system becomes an expert system.
One step in building an expert system is obtaining and encoding the
collective knowledge of human experts into the machine readable expert
system language. The specific implementation details of this encoding step
is largely a function of the particular syntax of the expert system
programming language selected. As an example one such expert system
programming language is PROLOG which is described in the text "Programming
In Prolog" by W. F. Clocksin and C. S. Mellish, Springer Verlag Inc., New
York, N.Y. (1981). Basically, the expert system contains a set of rules
and instructions on how the rules are to be applied to the available facts
to solve a specific problem.
In the intelligent storage manager 103, expert system techniques are used
to monitor the performance of the computer system memory. The
Configuration module 172 is a key element of the intelligent storage
manager 103. The Configuration module 172 analyzes the system
configuration of memory devices based on definitions in the operating
system 102 on every computer in the complex(s) and queries sent to every
element in the configuration to determine nature. It builds a model
configuration which is presented to an administrator for review and
approval. If the administrator desires to change the configuration he/she
can do so thru an interactive computer session. The facts gathered through
this monitoring operation are then used to identify modifications to the
organization of the computer system memory as well as the data stored
therein to improve the performance of the computer system memory. FIG. 2
illustrates some of the various routines or subsystems that are
implemented in intelligent storage manager 103.
Included are cache volume creation function 105, and DASD performance
optimizer function 106. Cache subsystems often end up in a state where
there are no volumes on the DASD subsystem suitable for caching, or where
the number of I/Os to cached volumes account for only a fraction of the
total number of I/Os to the data storage subsystem. The cache volume
creation function 105 creates volumes (205) on the data storage subsystem
that are suitable for caching. Cache volume creation 105 monitors (201)
the performance of the cache memory 160 and identifies where cache
| | |
