Indias parallel computers, Computer Engineering

Assignment Help:

Parallel Computers

In India, the design and development of parallel computers in progress in the early 80?s. The Indian Government recognized the Centre for Development of Advanced Computing (CDAC) in 1988 with the endeavor of building high-speed parallel machines. CDAC built and designed a 256 processors computer with INMOS T8000 series processors in 1991.  The other collection which developed parallel machines were at Centre for enlargement of Bhabha Atomic Research Centre, Telematics, Defence Research, Indian Institute for Sciences and Development Organization. The systems organized by these organizations are supposed to be the state of the art of parallel computers. It is generally decided that all the computers built by 2020 will be inherently parallel.

Next, we enumerate most important features of various generations of parallel systems developed in India.

Salient Features of PARAM series:

PARAM 8000 CDAC 1991:  INMOS 8000 transputer,256 Processor parallel computer as processing element. Peak concert of 1 Gigaflop. Application software weak.

PARAM 8600 CDAC 1994: PARAM 8000 better with Intel i860 vector microprocessor.  One vector processor for 4 INMOS 8000. Vectorized Fortran. Enhanced software for numerical applications.

 

PARAM 9000/SS CDAC 1996 : Used Sunsparc II processors and an interconnection switch made of INMOS transputers.

Important Features of MARK Series:

Flosolver Mark I NAL 1986: Used 4 Intel 8086 processors with 8087 co-processors. Proof of concept design.

Flosolver Mark II NAL 1988: 16 Intel 80386 processor and 80387 floating-point processor connected to Multibus II backplane bus for interprocessor communication. Used for solving fluid dynamics problems using Fortran.

Flosolver Mark III NAL 1991:  8 Intel i860 vector processors associated using message passing co-processor on a back plane bus. i860 were rated at 80 Mflops peak.  Fluid dynamics Codes(FDC) were optimized for the architecture.

Important Features of ANUPAM Series:

ANUPAM Model 1 BARC 1993: 8 Intel i860 processors connected to a Multibus II.Eight such clusters connected by using 16-bit SCSI interface. Used Front-end processor to assign responsibilities to the parallel computing cluster.  One parallel program at a time could be run.  Fortran environment.

ANUPAM Model 2 BARC 1997:  DEC Alpha processors associated by ATM switch in a cluster. DEC Unix environment.  High concert Fortran compiler to run data parallel programs.


Related Discussions:- Indias parallel computers

Explain direct addressing mode with example, Q. Explain Direct Addressing M...

Q. Explain Direct Addressing Mode with example? Direct Addressing Mode A direct operand signifies to contents of memory at an address referred by the name of the variable.

Explain extended binary coded decimal interchange code, Q. Explain Extended...

Q. Explain Extended Binary Coded Decimal Interchange Code? Extended Binary Coded Decimal Interchange Code (EBCDIC) is a character-encoding format employed by IBM mainframes. It

Machine centred versus human centred, Machine Centred versus human Centred ...

Machine Centred versus human Centred The discussion here is based on the difference in approach to the design of the work system when we prioritise either the needs of the mac

Decomposition model of parallel programming, The PVM system supports functi...

The PVM system supports functional and data decomposition model of parallel programming. It attaches with C, C++, and FORTRAN. The C and C++ language bindings for the PVM user inte

Define memory cycle time, Define memory cycle time? It is the time dela...

Define memory cycle time? It is the time delay needed between the initiations of two successive memory operations. Eg. The time among two successive read operations.

Explain briefly how firewalls protect network, Explain briefly how firewall...

Explain briefly how firewalls protect network.       A firewall is simply a program or hardware device that filters the information coming by the Internet connection into your

Use of large register file, Generally the register storage is faster than c...

Generally the register storage is faster than cache andmain memory. Also register addressing uses much shorter addresses than addresses for cache and main memory. Though the number

Which class used for multicasting in ip addressing scheme, In IP addressing...

In IP addressing scheme, class used for multicasting is: A class used for multicasting in IP addressing scheme is class D.

And-introduction rule, And-Introduction: In generally English says tha...

And-Introduction: In generally English says that "if we know that a lot of things are true, so we know that the conjunction of all of them is true", then we can introduce conj

Discuss the concept of segmentation, Discuss the concept of segmentation? ...

Discuss the concept of segmentation? Segmentation is method for the non contiguous storage allocation. This is different from paging as this supports user's view of his program

Write Your Message!

Captcha
Free Assignment Quote

Assured A++ Grade

Get guaranteed satisfaction & time on delivery in every assignment order you paid with us! We ensure premium quality solution document along with free turntin report!

All rights reserved! Copyrights ©2019-2020 ExpertsMind IT Educational Pvt Ltd