Table of Contents
| 1. |
Q. |
What is SolarisTM 9 Resource Manager
? |
| |
| |
A. |
SolarisTM 9 Resource Manager is a set of resource management
and network quality of service features that enables customers the
means to allocate and control resources such as CPU, physical
memory, and network I/O bandwidth. Solaris 9 Resource Manager is
integrated into the Solaris 9 Operating Environment (OE) kernel and
provides the framework for controlling system and network resources
for users, groups, or applications to provide more predictable
service levels. |
| |
Back
to top
| |
A. |
Sun plans to roll out components of Solaris containers in phases.
The release of Solaris 9 Resource Manager marks the first step. It
provides the framework for Solaris containers. With Solaris 9
Resource Manager, system administrators can establish resource
boundaries (also known as resource pools) for a specific
application, eliminating competition of resources with other
applications. The first resource boundary which system
administrators can establish with Solaris 9 Resource Manager is CPU.
In future phases, system administrators will be able to establish
boundaries for physical memory, swap space, and network I/ O
bandwidth. |
| |
Back
to top
| |
A. |
Solaris containers are Sun's next advance in server
virtualization. Solaris containers isolate software applications or
services using flexible, software-defined boundaries. Solaris
containers create computing environments within a single instance of
the Solaris Operating Environment and provide full resource
containment, fault isolation, and security isolation. The Solaris
containers will become the fundamental, ubiquitous management object
in the Solaris Operating Environment, and will be used throughout
Sun's entire product line. |
| |
Back
to top
| |
A. |
Solaris Resource Manager 1.2 and Solaris Bandwidth Manager 1.6
are unbundled products for controlling system and network resources,
respectively, for the Solaris 2.6, 7 or 8 Operating
Environments. Solaris 9 Resource Manager, on the other hand, is
integrated into the Solaris 9 kernel. While Solaris 9 Resource
Manager contains features similar to what is delivered in Solaris
Resource Manager 1.2 and Solaris Bandwidth Manager 1.6, the code
base of Solaris 9 Resource Manager is completely different.
Moreover, Solaris 9 Resource Manager also contains features which
are not part of the existing unbundled products. |
| |
Back
to top
| |
A. |
Solaris 9 Resource Manager is essential for customers who are
pursuing a server consolidation strategy to reduce costs through
simplifying their environments into fewer servers. This capability
provides mainframe like resource management in the networked world.
In addition, Solaris 9 Resource Manager allows Service Providers or
hosting companies to charge for multiple e-commerce sites on a
single server. Customers can gain economies of scale in management
and greater efficiency by using larger servers instead of a number
of small ones. |
| |
Back
to top
| |
A. |
Integration with Solaris Operating Environment The features of
Solaris 9 Resource Manager are integrated into the Solaris Operating
Environment. This tight integration enhances the manageability,
stability, and accounting capabilities for Solaris 9 Resource
Manager. By integrating into the Solaris OE kernel, it also enables
Solaris 9 Resource Manager to better support other core Solaris OE
features. Moreover, integrating Solaris 9 Resource Manager into the
Solaris OE provides the foundation for the continued delivery of
powerful resource management features in future releases. |
| |
Back
to top
| |
A. |
Solaris 9 Resource Manager provides effective resource control in
the following areas:
- Server consolidation
- Web hosting
- Large or varied user populations
- High Performance Computing Environments
|
| |
Back
to top
| |
A. |
Generally, the performance overhead is very low since Solaris 9
Resource Manager is tightly integrated with the core OS. System
administrators can define a policy for a group of processes in
contrast to the one-on-one nature of the TS scheduler. |
| |
Back
to top
| |
A. |
Yes, you can establish upper bounds for CPU resources. You can
dedicate one or more CPUs (CPU set) to a specific application
service. Only applications or users that belong to this specific
application service are allowed to use this CPU set. |
| |
Back
to top
| |
A. |
FSS enables you to control CPU resources to projects based on
their relative importance. |
| |
Back
to top
| |
A. |
In addition to being more dynamic, shares give more flexibility
than fixed percentages. Instead of allocating a fixed amount of CPU
resources to projects, the scheduler allocates the CPU resources
based on a proportion: the amount of the system the project is
entitled to (based on shares), compared to the total amount of
shares assigned to the currently active projects. It is also easier
to change shares, since the sum of shares doesn't need to add up to
a fixed percentage. Everything is relative, which makes scripting a
lot easier. |
| |
Back
to top
| |
A. |
Integration with Projects Previous version of Solaris
Resource Manager allocated CPU time on a per-lnode basis, as shares
of the processors on the system. In Solaris 9 Resource Manager, the
fair share scheduler is integrated into Solaris 9 kernel. Resources
will be allocated on a per-project basis and stored in a project
database, which can be a local file, or a NIA or LDAP database on a
central server.
Flattening the Share Hierarchy Previous versions of
Solaris Resource Manager allowed a system administrator to create
an lnode hierarchy. The hierarchy structure can be difficult to
administer and manage, and adds overhead to the implementation due
to the need to continuously recalculate shares in the hierarchy.
In Solaris 9 Resource Manager, the hierarchical structure is
flattened. Shares will be listed as a proportion of the entire
system, making the relative share of each project to be more
visible. Moreover, the flat structure makes it easier to manage
sites with multiple systems.
Integration with CPU Set In the previous version of
Solaris Resource Manager, the allocation of CPU shares to users was
performed without attention to the CPU sets in which the users'
processes were running. The result was often false and confusing. In
Solaris 9 Resource Manager, the fair share scheduler was revised so
that processes in each CPU set are treated independently - 'shares'
are counted locally for an individual CPU set. Moreover, the CPU set
configuration is now persistent over reboots. |
| |
Back
to top
| |
A. |
The project is an administrative identifier used to classify
related work. The project identifier can be thought of as a workload
tag equivalent to the user and group identifiers. The project can be
used to represent the workloads in which the user (or group of
users) is allowed to participate. A project could represent a single
user, a department, or an instance of a database. |
| |
Back
to top
| |
A. |
Yes, a user (or group of users) can belong to one or more
projects. A user will have a default project assigned. However, the
processes that the user launches can be associated with any of the
projects of which the user is a member. This means that a user can
have processes running in different projects simultaneously,
depending on the type of work the processes are doing. |
| |
Back
to top
| |
A. |
The task is a collection of processes that represents a set of
work within a project. A task can also be viewed as a workload
component. For instance, a task can represent a query to a specific
database instance. |
| |
Back
to top
| |
A. |
Tasks are created at login or by using the newtask(1) command. |
| |
Back
to top
| |
A. |
A project is a tag that associates similar work, whereas, a task is a job that is done within the project. Each task is associated with
one project. For instance, a project can represent a specific
database instance while a task can represent a query to that
specific database instance. |
| |
Back
to top
| |
A. |
The resource configuration information associated with a project
and can be stored in the project name service database which can be
a local file, or a NIS or LDAP database on a central server.
|
| |
Back
to top
| |
A. |
When a resource limit is reached, a handler is called. This might
return an error preventing the action causing the resource limit to
be exceeded, or send a message to the console or a user process, or
simply record the event and continue, depending on how the system
administrator configured the system. |
| |
Back
to top
| |
A. |
Solaris 9 Resource Manager features which are available in the
Solaris 9 OE when it ships will be included at no additional cost.
Customers do not need a separate license to use Solaris 9 Resource
Manager. However, customers may be required to pay an additional
licensing fee for Solaris 9 Resource Manager features which are
shipping in subsequent Solaris 9 update releases. |
| |
Back
to top
| |
A. |
There will not be any new releases of these unbundled products.
Rather, much of the functionality available in these products will
be integrated into the Solaris 9 Operating Environment. Solaris
Resource Manager 1.2 and Solaris Bandwidth Manager 1.6 will continue
to be supported, and patches will be released as necessary to
provide bug fixes to the software for the Solaris 2.6, 7 or 8
OE. |
| |
Back
to top
| |
A. |
No. There is no specific date currently planned for discontinuing
Solaris Resource Manager 1.2 and Solaris Bandwidth Manager 1.6.
Solaris Resource Manager 1.2 and Solaris Bandwidth Manager 1.6
support Solaris 2.6, Solaris 7 and Solaris 8. We will continue to
support Solaris Resource Manager 1.2 and Solaris Bandwidth Manager
1.6 for at least as long as Sun is shipping and selling Solaris 2.6,
Solaris 7 and Solaris 8. |
| |
Back
to top
| |
A. |
Solaris 9 Resource Manager provides the same basic functionality
as Solaris Resource Manager & Solaris Bandwidth Manager, but
with some significant enhancements and some entirely new features.
Given these new features, and changes in resource administration and
control, some customers may see differences in behavior between the
products. For instance, though the behavior of the fair share
scheduler will remain essentially the same, the lnode hierarchy
structure will be replaced with a flat set of projects which do not
have any parent-child relationship. A migration script will be
available to allow customers to easily upgrade a system with Solaris
Resource Manager 1.x installed to one running Solaris 9 Resource
Manager without losing the Solaris Resource Manager 1.x
configuration information. Documentation will also be made available
to assist customers in converting from the hierarchical lnode
structure to the flat-based project structure. If the customers do
not depend on a multi-level lnode tree, they can deploy Solaris
Resource Manager today and adopt Solaris 9 Resource Manager on the
Solaris 9 OE in the future with little change in behavior.
Once the lnode configuration is converted to the new project
structure, customers can begin to take advantage of the new,
powerful features of the Solaris 9 Resource Manager such as resource
pools and the statistical and accounting tools. |
| |
Back
to top |
To learn more about the Solaris 9 Operating Environment, visit www.sun.com/solaris.
|