What I'm writing below is my attempt to summarize the web
pages (press release, overview white paper, demo, etc.), as
well as a short opinion of what they're missing.
For more information on the Web-based Enterprise Management effort,
visit its Web site -- http://wbem.freerange.com/
If you don't have time to check out my summary below, but still
want to know what this system can do, check out the cool demo:
http://www.compaq.com/productinfo/systems/strategy/urm/dmi_snmp/mgmtstrt.html
Better still, check out the demo of an implemented HMOM:
http://www.compaq.com/productinfo/systems/strategy/urm/my_computer/hmom_frame.html
This industry standard isn't a vaporeffort; it's actually working now.
The implications for things like hooking-devices-to-the-Internet bode
well. Our opinion of what they're missing (see Section 8, below)
has possible implications to the next generation of HTTP.
-- Adam
------------------------------------------------------------------------
1. THE PROBLEM BEING SOLVED.
Organizations want improved access to distributed resources, without the
burden of managing diverse systems. The systems available today for
managing heterogeneous networks have several major shortcomings:
A. The widely used approaches to network, systems, and applications
management have created incompatible infrastructures. These complex,
platform-dependent approaches are costly, since they require sys admins
to deal with multiple technologies.
B. Software developers must create applications for multiple (often
competing) management environments. Since they can't integrate their
applications with other solutions, their ability to provide innovative
solutions is restrained.
C. Current management platforms confine ability to scale as customer
needs grow.
D. They are centralized and do not take advantage of the powerful
technology they are designed to manage.
E. They compile data in inconsistent formats, complicating the
comparison of data from different management platforms -- and, as a
result, limiting the value of the data collected.
Working collectively to develop open management systems, users, vendors
and other industry constituencies have made limited progress. Despite
prolonged debate, they have failed to devise a viable solution. The
protocols and standards that have evolved -- SNMP and DMI, for example
-- have eased the task of managing individual resources, but the
problems posed by solution diversity continue to make enterprise
management extremely costly and complex.
------------------------------------------------------------------------
2. GLOSSARY OF JARGON.
Common Object Request Broker Architecture (CORBA)
An architecture specified by the Object Management Group that supports
interoperable, distributed object oriented systems.
Desktop Management Interface (DMI)
A standard management protocol designed and administered by the DMTF.
Desktop Management Task Force (DMTF)
An industry organization that manages the DMI specification.
Distributed Component Object Model (DCOM)
Microsoft's architecture for use of objects in a distributed enterprise.
HyperMedia Managed Object (HMMO)
A managed entity that has data that can be either interrogated or
managed by a browser, either directly or through a management schema.
Every framework object has at least one URL.
HyperMedia Object Manager (HMOM)
An application that provides hierarchical control point for accessing
and managing other HMMOs on the network, services to manage large
numbers of managed objects, gateway agents to map HTTP request to the
native protocol of the non-HMMO entity, such as SNMP and DMI.
HyperMedia Management Protocol (HMMP)
An object-oriented management protocol implemented on top of HTTP.
HyperMedia Management Schema (HMMS)
An extendible, object oriented data model that is used to model
the managed environment.
Hyper Text Markup Language (HTML)
A language used to create documents that are processed and displayed
by Web browsers.
Hyper Text Transfer Protocol (HTTP)
The data transfer protocol most associated with Web-based
communications and information sharing.
Internet Engineering Task Force (IETF)
A standards body whose focus is on protocols used on the Internet.
Managed Object
A system component that provides information required for management
purposes.
Simple Network Management Protocol (SNMP)
A standard network management protocol developed for the
TCP/IP Internet to communicate between agents and managers
running diverse platforms.
Web-based Enterprise Management
Grass-roots efforts within the management community to leverage Web
technology, especially browsers, into existing products.
------------------------------------------------------------------------
3. OVERVIEW OF THE WEB-BASED STANDARDS EFFORT.
On July 17, 1996, BMC Software, Cisco, Compaq, Intel, and Microsoft
proposed an industry standards effort that will allow administrators to
use any Web browser to manage disparate systems, networks, and
applications. The intent of the Web-Based Enterprise Management effort
is to enable the development of tools that reduce the complexity and
costs of enterprise management.
The Web-Based Enterprise Management standards effort is open to the
entire hardware, software, OEM and internetworking community, many of
whose members will announce support for the effort today. The effort is
designed to integrate existing standards, such as
A. Desktop Management Interface (DMI/RPC) for desktops and servers;
B. Simple Network Management Protocol (SNMP/UDP) for networks; and
C. Hypertext Transfer Protocol (HTML/HTTP) for Internet communication.
These standards will be aggregated into an architecture that can be
managed using any Web browser. Actually, these standards WERE
integrated already, using existing technologies; their demo works
fine. I think they could use an official flagship like W3C to promote
this effort, but we'll see how it goes. Meanwhile, I'll continue
to talk in the future tense of what they plan to do.
The initial proponents of this industrywide effort agreed to cooperate
on developing resulting standards, and none of them will own the
standard. They've learned from past mistakes; previous industry
attempts at integrating enterprise management have resulted in a variety
of frameworks and incompatible management protocols and data models.
Since many customers have already invested heavily in SNMP-enabled and
DMI-enabled systems and devices, and since the Web technology is so
promising, this solution makes sense from the interoperability and
scalability standpoints.
------------------------------------------------------------------------
4. NEW TECHNOLOGIES.
Structured and unstructured data can be presented by some combination of
DMI/RPC, SNMP/UDP, and HTML/HTTP. Two new management-related
technologies provide data modeling, manipulation, and communication
capabilities, as outlined recently at meetings of the DMTF and IETF:
A. HyperMedia Management Schema (HMMS), an extensible data model
representing the managed environment; and
B. HyperMedia Management Protocol (HMMP), a communication protocol
embodying HMMS, to run over HTTP.
In addition, a portable C++ implementation of a HyperMedia Object
Manager (HMOM) will also be created, with plans to make it freely
available. HMOM will manage elements as objects, integrating management
data and coordinating control through a variety of management protocols
and interfaces (such as SNMP, DMI, etc.) through Web-based consoles and
other management applications. The specification for HMOM and the
portable C++ implementation will be placed in the public domain.
------------------------------------------------------------------------
5. KEY ADVANTAGES.
The standards effort provides three key advantages:
A. SCALABILITY. With a simple Web browser as the management
interface, organizations can cost-effectively take advantage of
networking technology they already have in place to manage a wide range
of network resources, such as routers, hubs, PCs, workstations,
distributed applications, and databases. When the same technologies
used for building networks are used to create management applications,
the scalability of the applications can match that of the network.
The proposed standards provide such scalability by allowing a system
administrator to learn and implement just one interface to monitor and
maintain low-end devices and systems as well as mainframes and
everything in between. The standards will support a broad range of
management solutions and will build on Internet innovations to
meet the demanding requirements of the most complex heterogeneous
computing environments.
B. INCREASED CHOICE IN APPLICATIONS, GREATER FUNCTIONALITY.
The proposed open standards offer a single foundation on which to build
management applications, obviating the need to design different versions
for different management platforms and making applications more
efficient and cost-effective. They thus free developers to concentrate
on innovative functionality rather than system differences and allow
them to bring applications to market more quickly. A major benefit for
users will be significant: a greater selection of management
applications and added functionality that takes advantage of rapidly
evolving Web technology.
C. LOWERED COSTS FOR SET UP AND OPERATION. A single interface for
managing all networks, systems and applications will greatly reduce the
complexity that currently frustrates system administrators. The proposed
management standards will free them from having to access management
applications from specially outfitted consoles. Instead, they will work
at any Web-enabled client systems distributed throughout an organization
to access distributed management applications. Access is controlled by
the security measures implemented within HTTP. A management system based
on a Web browser interface that eases access to management data for
networks of UNIX, Windows NT, MVS, VMS, and Netware platforms will be
less costly to learn, set up, operate, and support. Likewise, as easier
application development is enabled, management solutions will
proliferate and competition will impact prices. Today thousands of
developers are creating open Internet solutions. The Web-based
technology will enable them to apply this innovation to distributed
management applications.
HTML applications will link users to the rich resources of the Internet.
As a result, applications built on the proposed standards will be able
to connect system administrators and less knowledgeable users to vendor
Web sites and other locations, where they can find timely information on
the applications they depend on. For example, at the Web site of an
independent software vendor, users could consult Help bulletin boards
and product documentation illustrated with 3D drawings, instructional
audio programs, tutorial text, and other materials would guide them
step-by-step through their management applications. The flexibility of
the Internet would allow for interactive, timely information that
proposes customized courses of action.
As more business-process data in corporations becomes accessible via a
Web browser, system administrators will access management information
and operational information with that browser. As a result, using the
same interface that makes corporate information available, a system
administrator could, for example, check the accounting Web server to
determine the best time to schedule downloads.
------------------------------------------------------------------------
6. A DEMONSTRATION OF THIS SYSTEM'S POWER.
In its simplest form, the Web-Based Enterprise Management initiative
defines management as a Web browser linking directly to a device or
application (called HyperMedia Managed Object, HMMO) that provides
management information in an HTML page.
Direct connection of a Web browser to a HMMO represents the ultimate in
low-end scalability. The user is no longer concerned about the
underlying management protocols employed on the managed object, so long
as there is an HTML interface exported on the device (i.e., a management
Home Page).
Any device on a network accessible via the HTTP protocol is considered a
HyperMedia Managed Object (HMMO).
Objects within a device (e.g. subsystems, operating system,
applications) are further examples of HMMO's. These "internal" objects
expose their attributes through the device's home page.
Managed Objects may wish to expose a simple hyperlinked home page
offering various views of their available management data:
A. Simple hyperlinked text view of management data when
network bandwidth is an issue.
B. Media rich content if presentation and organization of
data outweighs network bandwidth concerns.
A device may expose embedded managed objects through its home page.
Example home page has hyperlinks to managed objects:
MY COMPUTER HOME PAGE
* Managed Objects:
o Compaq Deskpro
o Compaq QV210 Monitor
o HP Printer
o Microsoft Windows NT, v3.51
o Internet Server
The HyperMedia Management Application (HMMA) is a system component whose
primary function is to provide monitoring/management information to a
browser, and to accept control commands from the browser.
A. Aggregates devices, both HMMO and non-HMMO (SNMP and DMI).
B. Responds as a HTTP server and provides management information and
control options.
http://www.compaq.com/productinfo/systems/strategy/urm/hmom_frame_gl.html
presents an enterprise level system management solution. Check it out,
the graphics are pretty neat.
In this example, an HMMA presents an upper level (global) view of
manageable sites within the enterprise.
A. Links are provided directly to scattered geographical sites and to
localized views within those sites.
B. Alerts have been appropriately filtered as they progress up to this
global view.
C. The user is hyperlinked to detailed information for each alert.
D. A search/filtering button is provided to leverage Web based discovery
and search tools.
Now we consider the site map view of the HMMA. The user has selected
view of a specific geographical site.
A. Links are provided back to a global view (if available) and
individual workgroup views (Engineering, Manufacturing, etc.).
B. Alerts have been appropriately filtered from each workgroup and
managed object.
C. The user is hyperlinked to detailed information for each alert.
D. A search/filtering button is provided to leverage WEB based
discovery & search tools.
Now we consider the workgroup map view of the HMMA. The user has
selected view of a specific workgroup.
A. Links are provided back to a global view and site view (if
available). A button provides the to/from "list view" (text based
hyperlink view).
B. Alerts have been appropriately filtered from each individual
managed object.
C. The user is hyperlinked to detailed information for each alert.
D. A search/filtering button is provided to update map view since
last network scan.
E. The user is hyperlinked directly to device (HMMO) home page when
selected in the map frame.
Now we consider an example computer home page of the HMMO.
A. The user has worked down to view of a specific device/HMMO.
B. This example home page utilizes rich media and provides overview
information about the HMMO.
C. Buttons are provided to detailed management/control data and
change/history logs.
D. Alerts frame provides hyperlinks to detailed information for each
alert.
Note that the HMMA may act as a gateway to provide support for large
installed base of non-HMMO (SNMP/DMI) devices. This preserves the
current investment in SNMP/DMI managed devices. The HMMA may integrate
data from non-HMMO's in unified object maps.
This may have flown past you without sinking in; in that case, I suggest
you play with the Web demo they've set up. The cool thing is, it really
works! It's quick, it's intuitive, and it's the right solution to the
problem:
http://www.compaq.com/productinfo/systems/strategy/urm/dmi_snmp/mgmtstrt.html
Better still, check out the demo of an implemented HMOM:
http://www.compaq.com/productinfo/systems/strategy/urm/my_computer/hmom_frame.html
------------------------------------------------------------------------
7. WHAT'S NEXT?
BMC Software, Cisco, Compaq, Intel, and Microsoft will promote the
adoption of the proposed standards in several ways. Working together,
and with other organizations that support the effort, they will
communicate the benefits of a Web-browser-based framework to the
industry. The five sponsors have invited other parties interested in
easing the cost and complexity of network, applications and systems
management to participate in the effort.
Components of this proposed standard have already been offered to
industry standards committees to help promote them in the public forum.
The HMMP will be submitted to the IETF. The HMMS will be presented to
the DMTF. A reference implementation of an HMOM will be developed and
placed in the public domain to initiate development in the broader ISV
community. Together, vendors, users, and standards bodies can move
network and systems management a major step forward.
For more information on the Web-based Enterprise Management effort,
visit its Web site -- http://wbem.freerange.com/
------------------------------------------------------------------------
8. WHAT THIS TECHNOLOGY REALLY NEEDS
(this is the opinion of r.k. and a.r., mind you)
This system could really become powerful, if they add two notions:
A. TRANSACTIONALITY. By transactionality, I mean that any system
interactions should be operations with one of two outcomes: commit or
abort. Transactions with the system are recoverable, replicable,
time-stampable, and secure. The transactions are also be nestable in a
manner that preserves consistency of operations.
B. AUTOMATABILITY. Right now, a human operator is required to make
use of this system. But one of the tremendous powers of the Web (and
especially the next generation of HTTP) is the concept of the ability to
automate transaction operations. If you looked at the demo, think how
great it would be if we could have machines in front of those forms and
widgets, instead of a human, to do things like fault-tolerance, logging,
system checking, and automation of tasks.
It's unclear that the industry standards board is working on either
of these, or even that they should. Perhaps both of these concepts
should just be added to the current incarnation of HTTP, and that
solution would propogate to other standards such as the one in this memo.
------------------------------------------------------------------------