Previous section.
UMA Measurement Layer Interface (MLI)
Copyright © 1997 The Open Group
Introduction
Purpose
This document is one of a family of documents that comprise the
Universal Measurement Architecture (UMA), which define
interfaces and data formats for Performance Measurement.
UMA was originally defined by the Performance Management Working Group (PMWG)
and subsequently adopted by The Open Group.
This document is a specification for the functional
characteristics of the Measurement Layer Interface (MLI), as
defined in the UMA Reference Model in
UMA Reference Model
.
The document describes the
underlying semantics and the function calls that implement
them. It also defines a format for headers appended to data as
captured by the low-level Data Capture Interface (DCI). (These headers are
part of the control and status messages sent between UMA and the
applications it serves.)
The UMA is defined in the following documents:
-
Universal Measurement Architecture Guide (see reference UMA).
This document provides an overview of the UMA.
-
UMA Measurement Layer Interface Specification (this document).
-
UMA Data Capture Interface Specification (see reference DCI).
This document defines a standard programming interface for capturing
system and application provided data.
-
UMA Data Pool Definitions (see reference DPD). This document defines a
performance data pool for the analysis and management of computer systems,
and an organisation to facilitate the collection and use of such data.
Scope of the Universal Measurement Architecture
The scope of the
Universal Measurement Architecture is:
-
to provide a set of common measurement control and data
delivery services for (client) performance applications
-
to provide seamless access to current and historical
measurement data
-
to insulate the operating system kernel from performance display and
analysis applications by means of a common application programming
interface (API)
-
to maintain portability of user tools to any systems that
implement the architecture (again through the common API)
-
to provide specific mechanisms for data capture that implement
metric registration functions from distinct data providers
-
to provide a mechanism for control of the instrumentation that
coordinates the capture of kernel and non-kernel data sources
-
to support access to distributed performance functions and data from
remote nodes.
The UMA architecture specifies two interfaces:
-
the MLI -
a high-level application programming interface for
specification and reporting of formatted measurement data,
(this document)
and
-
the DCI - a low-level application programming interface for acquisition of
raw kernel, trace, event, and subsystem data,
(see reference DCI).
In addition, the architecture provides services for distributing data to multiple
applications, for maintaining historical data,
and for synchronising the capture of metrics from different sources.
Performance and capacity management of operating
systems have been considered
internal
to the operating system and as such differ from one operating system
to another operating system,
and from one implementation to another implementation. Most
operating systems have, as a matter of necessity,
performance analysis modules, narrowly targeted at
the type of hardware, software and networking
facilities implemented within the system.
Most operating systems provide ad-hoc developed or
tailored performance metrics. Some of these tools
are developed as internal support tools for
benchmarking or on demand of performance analysts
and capacity planners. These tools are generally
also confined to one machine only and can not be
interrogated remotely.
The new era of networking and interoperability
views performance management and capacity planning
from the user's perspective. Multiple machines and
operating systems can be involved in the
interaction with the user. This approach requires
capture and presentation of performance metrics to
be clearly defined and portable between platforms
and operating systems.
Definitions, Acronyms and Abbreviations
Terms, acronyms and abbreviations used in this specification
are defined in the Glossary.
Conformance
A conformant implementation must support all of
normative requirements in this MLI
specification, that is, as specified in chapters 1 - 8)
except in the following respects:
-
Support of unsolicited events:
MLI support of unsolicited events is optional.
When a measurement application requests delivery
of unsolicited events from an MLI implementation
which does not support this feature, status and
reason codes will be returned from the call
indicating the condition.
-
Support of metric, instance tag and work unit description by metadata:
MLI support of metric, instance tag and work unit
description by metadata is optional. This
information, when available, is presented
in the "Subclass Attributes" subclass of the
class "UMA Configuration" solicited with the
umaRequestConfig()
MLI call. When not supported,
an attempt to retrieve this subclass will return
status and reason codes indicating that the
subclass is not implemented.
-
Support of dynamic data:
MLI support of dynamic data is optional.
The availability of dynamic classes and subclasses
is indicated in the subclass "Implementation"
of class "UMA Configuration" by the setting of
the bit flag UMA_DYNAMIC for each class and
subclass. Support of dynamic data implies the
MLI support of metric metadata.
-
Support of protocol section contents in control messages:
The only required usage of the control UDU header
protocol section is that the
umaCreate()
and
umaReconnect()
logical message protocols and their acknowledgements include the protocol
section field describing the sending platform's
wordsize (mh_wdsize). The other fields may
be used to support private protocols between MLI service
layers. The presence or absence of the additional
protocol fields has no impact on the use of the MLI API.
-
MLI use of the DCI:
Use of the DCI by the MLI for data acquisition
is optional. However, if the DCI is used as a
source of data, the user should refer to the
DCI conformance requirements.
-
Enhanced security services:
These are specified external to the MLI.
-
WorkInfo granularities:
While MLI implementation of the Workload
definition mechanism (using UMAWorkInfo) is
mandatory, the availability of specific WorkInfo
granularities is defined by the data provider.
Why not acquire a nicely bound hard copy?
Click here to return to the publication details or order a copy
of this publication.