IBM TotalStorage Productivity Center V3.1: The Next Generation

Chapter 2. Key concepts 37

2.7.1 Collecting data in general

IBM TotalStorage Productivity Center uses multiple techniques to collect data from the

various subsytems and devices in your environment. Figure 2-14 shows presents these

various data paths where data is coming from and passing across TotalStorage Productivity

Center. The data paths are:

 CIM Agents

 The IBM TotalStorage Productivity Center Data and Fabric Agents

 SNMP Traps and Events

 StorageManagement Interface-Specification (SMI-S) queries

 API calls through vendor specific product APIs

Figure 2-14 Data paths within the TotalStorage Productivity Center environment

These data paths can be used on demand to collect data. In this case the queries are

triggered by manual interaction with the system. But in general you will setup your

TotalStorage Productivity Center environment to perform data collection at intervals you

define. Each time when these jobs are run, another record of information is stored in the

associated database tables. The continuously gathered information is the basis for all reports,

statistics, performance charts and other reports that can be produced with TotalStorage

Productivity Center. The frequency of this data collection is a critical factor for the timeliness

of the data which is displayed in other elements of the GUI. This brings up the crucial fact,

that elements that are shown in the explorer - the elements you might want to perform a

specific action on - might be missing until you go out for the next data collection and update

the database contents.

The behavior of the Topology Viewer is to look up the information in the database instead of

collecting it from the managed systems. The principle with updating the database contents is

that only the scheduled jobs, scans, or probes will update the information.

c e n t r a l

T P C

D B

T P C

f o r D i s k

T P C

f o r D a t a

T P C

f o r F a b r i c

Data Agent

Fabric Agent

Data Agent

Fabric Agent

Proprietary

API

SNMP

Proprietary

API

SNMP

C I M

A g e n t

T o p o l o g y V i e w e r

Storage and Tape

Subsystems

Comput ers

SAN Components

38 IBM TotalStorage Productivity Center: The Next Generation

There are only two exceptions:

 A subsystem can issue an indication to its associated CIM Agent to have it update a

specific information in the database (for example, the addition of a newly created LUN).

 A device managed through SNMP ( for example, a switch) can send a trap indicating there

was a change to which TotalStorage Productivity Center should be notified.

In this version of the product, the information is stored in a single database. There is a second

database for Tivoli Agent Manager which is described this later in the book.

Data correlation within IBM TotalStorage Productivity Center

The design of the database and its tables are the basis for the information you can derive

from IBM TotalStorage Productivity Center. The logical relation - in terms of how a storage

infrastructure is functioning - between single entries in the database makes it possible to tell t

the condition and state of the subsystems and its counterparts to the left and to the right.

Startup of the main user interface

After installation of the product the user will be presented with the main GUI. The GUI holds

several containers which are filled up with items related to the following tasks. The next

paragraph tries to explain these tasks in brief fashion. Later in this redbook more details are

provided.

Figure 2-15 The main GUI window