RAC Grid Computing Fundamentals
CPU cycles, or more appropriately the idle CPU cycles, are in
demand. In addition, the CPU cycles are available in the servers, which are
under-utilized. For processing a large problem in a fairly small amount of time,
more CPU cycles are needed. The availability of such resources concurrently (in
parallel mode) helps to compute much faster than a single server and its
The next important component is the availability of
the data. Where is the data to process and then provide the results? The data
is stored in organized and structured databases. While storage systems
physically store the data blocks on media devices such as disk drives in the
storage arrays, provided mostly notably by, EMC, Hitachi and IBM, the servers
fetch and process in the name of relational databases.
The two important resources processing power and storage systems are the main areas of
attention for grid planners and grid users.
What is Oracle Grid Computing?
Back in 1998, Carl Kesselman and Ian
Foster in the book The Grid: Blueprint for a New Computing Infrastructure,
while attempting to give a broader vision of Grid, wrote:
A computational grid is a hardware and software infrastructure that provides
dependable, consistent, pervasive, and inexpensive access to high-end
Subsequently, Ian Foster with Steve Tuecke, redefined the definition stating that Grid Computing is concerned with
coordinated source sharing and problem solving in dynamic, multi-institutional virtual organizations.
They further noted:
The sharing that we are concerned with is not primarily file exchange but rather direct access to
computers, software, data, and other resources, as is required by a range of collaborative problem
solving and resource-brokering strategies emerging in industry, science, and engineering. This sharing is,
necessarily, highly controlled, with resource providers and consumers defining clearly and
carefully just what is shared, who is allowed to share, and the conditions under which sharing occurs. A set
of individuals and/or institutions defined by such sharing rules form what we call a virtual organization.?
However, from the client, user, or consumer point of view, Grid computing is seen more
of a utility of computing. Users do not care where their problem is computed or
analyzed and where the data comes from. They are merely interested in getting
the results, and getting it done faster and cheaper. From the server side, grid
is all about the pooling resources, virtualization, and provisioning.
Simply put, Grid Computing is the pool of computers actively glued into a
virtual computer by the other related components such as middleware software,
interconnects, networking devices and storage units. It is distributed computing
taken to a higher evolutionary level. With standards being worked out for the
effective sharing of resources from the Grid Pool and with the proper security
access levels; Grid is a new class of infrastructure.
Based on the technological changes occurring in the contemporary period, Ian Foster gives an
interesting justification for developing and implementing Grid Computing. In his words:
The annual doubling of data storage capacity, as measured
in bits per unit area, has already reduced the cost of a terabyte disk farm to
less than $10 000. Anticipating that the trend will continue, the designers of
major physics experiments are planning petabyte data archives. Scientists who
create sequences of high-resolution simulations are also planning petabyte
Such large data volumes demand more from our analysis capabilities. Dramatic
improvements in microprocessor performance mean that the lowly desktop or laptop
is now a powerful computational engine. Nevertheless, computer power is falling behind storage.
By doubling "only" every 18 months or so, computer power takes five years to increase by a single
order of magnitude.
Assembling the computational resources needed for large-scale analysis at a
single location is becoming infeasible. The solution to these problems lies in
dramatic changes taking place in networking.