Hive Computing FAQ
What problem does Hive Computing solve?
When it comes to mission critical computing, businesses and
other organizations face intense pressure to do more with less.
On one hand, they must manage larger transaction volumes,
larger user populations, and larger data sets. They must do all of
this in an environment that demands a renewed appreciation for the
importance of reliability, fault tolerance, and disaster recovery.
On the other hand, they must satisfy these requirements in
a world of constrained resources.
It is no longer an option to throw expensive hardware and
people at problems.
The challenge they face is that, when it comes to platforms
for building mission critical applications, the world is fragmented.
Different products are designed to satisfy different sets of
requirements.
Mainframes are capable but are extremely expensive to
acquire and maintain. Fault tolerant computers are available and
predictable but are inflexible, expensive, and hard to maintain.
Application servers are more scalable and easier to use but are
fragile, expensive, and inflexible. Newer solutions like Linux
clusters and distributed supercomputers offer tremendous scalability
but are not built for business.
As a result, when building mission critical applications,
too often businesses and other organizations must make tradeoffs and
settle for solutions that satisfy only a subset of their needs.
The goal of Hive Computing is to eliminate the need for
such tradeoffs and give businesses the best of all worlds:
extraordinary levels of reliability at
a fraction of the cost of existing solutions like mainframes, fault
tolerant computers, and application servers.
What is different about Hive Computing?
Hive Computing is able
to deliver on this promise because it is based on three very
different assumptions. First, a Hive assumes the application is all
that matters. A self-organizing aspects of a Hive enable developers
to focus on the task at hand, not the complexities of the physical
environment. Second, a Hive, like TCP, recognizes that failure
happens. As a result, a Hive is self-healing and is able to deal
with failure, not fear it. Finally, a Hive assumes computers are
disposable. The self-maintaining capabilities of a Hive allow it to
be built from inexpensive, PC-grade components.
What are the benefits of Hive Computing?
Hive Computing enables
order of magnitude reductions in the time and cost of developing,
deploying, and maintaining mission critical applications. This is
due to the fact that a Hive is simultaneously...
- Survivable
- Scalable
- Affordable
How does
Hive Computing fit into the world of Grid Computing?
Hive Computing and grid computing are different but
complementary approaches to computing.
The goal of grid computing is to create a computing
power grid that allows people and organizations to link together
large numbers of powerful computing resources in order to solve
large (and often previously unsolvable) problems. As a result, grid
computing is concerned with issues like security, billing,
autonomous control, and administration.
In contrast, Hive Computing is concerned with
building extremely reliable and affordable computing resources that
can be linked together using a grid.
To follow the electrical power system analogy, if
grid computing is concerned with the design and management of the
electrical distribution system (e.g. power lines) then Hive
Computing is concerned with the design and management of power
plants.
How does
Hive Computing fit into the world of Web Services?
The web services vision is based on the idea that
people and computers will be able to access, over the Web, services
that perform a specific set of functions (e.g. a service that, when
given a stock symbol, will return the current price). We believe
that a Hive, because it allows you to quickly and easily develop
solutions that are both reliable and affordable, offers the best way
to host and deploy such services.
|
