Are there any parallel computing databases?

Are there any parallel computing databases?

Post by J. S. Jense » Thu, 25 Jun 1998 04:00:00



What databases exist that make use of clusters of machines running
parallel libraries (PVM?) or RPC contructs?  Any?

--
J. S. Jensen

http://www.paramin.com

 
 
 

Are there any parallel computing databases?

Post by Art S. Kage » Fri, 26 Jun 1998 04:00:00



> What databases exist that make use of clusters of machines running
> parallel libraries (PVM?) or RPC contructs?  Any?

Both Oracle and Informix have Clustered/loosely coupled MP versions,
but, do not use either PVM or RPC.  Oracle relies on its global lock
manager (I do not remember the proprietary feature name) and Informix
uses an heuristic handshaking protocol (which has no published feature
name).

Art S. Kagel

 
 
 

Are there any parallel computing databases?

Post by J. S. Jense » Sat, 27 Jun 1998 04:00:00




> > What databases exist that make use of clusters of machines running
> > parallel libraries (PVM?) or RPC contructs?  Any?

> Both Oracle and Informix have Clustered/loosely coupled MP versions,
> but, do not use either PVM or RPC.  Oracle relies on its global lock

The reason I ask is that I would like to do scalability tests on
exceptionally complex SQL statements.  Results would be relative, thus the
db doesn't particularly matter (at least too much :-).  If there were any
dbs employing PVM or RPC, then I could truly test with a large number of
machines.

I'll keep looking...

--
J. S. Jensen

http://www.paramin.com

 
 
 

Are there any parallel computing databases?

Post by Martin Sandma » Sat, 27 Jun 1998 04:00:00





> > > What databases exist that make use of clusters of machines running
> > > parallel libraries (PVM?) or RPC contructs?  Any?

> > Both Oracle and Informix have Clustered/loosely coupled MP versions,
> > but, do not use either PVM or RPC.  Oracle relies on its global lock

> The reason I ask is that I would like to do scalability tests on
> exceptionally complex SQL statements.  Results would be relative, thus the
> db doesn't particularly matter (at least too much :-).  If there were any
> dbs employing PVM or RPC, then I could truly test with a large number of
> machines.

> I'll keep looking...

> --
> J. S. Jensen

> http://www.paramin.com

The NCR Teradata is also a Massively Parallel Processing database
system but it, too, does not use PVM or RPC. The interconnect bus
is called the BYNET. At NCR we have done scalability tests with it
for many years, especially with the TPC-D benchmark. A recent press
release is attached.

        -- Martin

Press Release Follows:

NCR  SENDS  TPC-D  THROUGH  THE  ROOF  WITH  A  3 TB  TPC-D

On Friday, June 19, NCR published the first-ever TPC-D benchmark at 3
TB, breaking industry records for high volume decision support
benchmarking.

Performed on a 32-node 5150 Teradata system, the benchmark has set the
standard at 3 TB with these characteristics:

-- 32-node 5150, with 4 x 200 MHz CPUs, 1 GB memory per node



-- Disk Storage Ratio = 3.69
-- 3 streams in the Throughput Test
-- Full Data protection with RAID5

Notable about this new TPC-D are the following:

-- NCR/Teradata once again achieved a feat unreachable by any other
platform today.  Competitors who have managed to produce a 1 TB  TPC-D
benchmark have done so only by totally exhausting themselves, even on
their "futures" platforms.  This higher volume point is simply out of
reach for them.  

-- This 3 TB benchmark configuration (32 nodes, 128 x 200 MHz CPUs, 32
GB memory) was substantially less hardware than used by IBM in their
recent 1 TB results (48 nodes, 192 x 332 MHz CPUs, 144 GB memory),
underscoring Teradata's ability to perform well with minimal hardware.

-- This very low disk storage ratio of 3.69 includes complete data
protection and sets to rest any speculation that Teradata solutions have
to carry high disk overhead, as do our competitors'.

-- The 3 concurrent query streams executed in 11% less time than the
single stream would have, had the single query stream been run 3 times
serially.  This 11% scale-up substantiates the strong multi-user
capabilities that Teradata consistently delivers on large and on small
systems, at low or at high volumes, with few or with many users.
-- Single = 21710.6 sec
-- 3-stream = 58,000
-- Single times # of streams = 21710.6 * 3 = 65131.8
-- Multi-time / 3 serial single streams = 58,000 / 65131.8 = .89 or 89%

-- Scale-up = 100% - 89% = 11%

-- Comparing last year's 1 TB produced by NCR/Teradata, on a similarly
configured 32 node 5150 platform with this recent 3 TB, the QppD per
node performance has increased from 380 per node (12,149 / 32 = 380) up
to 451 per node (14,446 / 32 = 451).  As the same level of parallelism
was active on each node, and the hardware was equivalent, this 20%
increase comes primarily from Teradata database and application
improvements.

--  While other vendors cannot even take this giant step up, Teradata is
nowhere near exhaustion with this 3 TB benchmark.   Both the hardware
platform (which can easily scale to 96 nodes and beyond, as we have seen
in production Teradata systems), and the Teradata database have been
proven very capable of taking on greater and larger challenges.  Nor was
the system under any particular stress during this execution.  

The good news for customers is the NCR/Teradata platform that produced
this 3 TB benchmark is available today, and was composed of current
hardware and software.  Further, it demonstrates a low disk storage
ratio and the same simple 2-page database setup representative of
Teradata systems around the world.

 
 
 

1. Parallel computing

Our research group has 3 RS/6000's (two model 350s and a 32H) that
I'd like to set up to allow us to do parallel computing, like one
can do on the SP/2 here at Cornell.  Does anyone know if something
like this is possible ?  If so, what software is required and
where do I get it ?

Thanks,
        Jeff

2. kerneld patches for 1.3.45

3. Intro to Distributed Memory Parallel Computing Workshop

4. Stealth 64 VRAM vs DRAM

5. q:s about parallel computing?

6. Can you install Linux on a Microchannel (MCA) machine?

7. Parallel Computing

8. HylaFAX and Page Chopping

9. !!!PARALLEL COMPUTING (DCE) ON LINUX. IS IT REAL???

10. Parallel computing question

11. Parallel Computing using SCSI?

12. Parallel computing using SCSI. (again, but different)

13. Parallel Computing/Clustering