Exascale computing refers to systems capable of performing 1 quintillion (1018) calculations per second, which is 1000 times more powerful than the first petascale computers. The goals of exascale are to develop systems by 2018 that can perform simulations with greater precision and speed. Such systems will require new architectures and memory designs to handle the power and reliability challenges at this scale. GPUs and new low-power processors will be important for exascale computing.

1. By
Balasudharsanakrishnan.B BE(CSE)
Shahanas Banu.T BE(CSE)
Pre –Final Year

2. INTRODUCTION
• Exascale computing refers to computing systems capable of billion
calculations per second.
• Such capacity represents a thousand time greater than the first
petascale computer that came into operation in 2008.
• One exaflops is a thousand petaflops or a quintillion 1018, floating
point operations per second.

3. ORIGIN
• At a supercomputing conference in 2009, Computerworld
projected exascale implementation by 2018.
• In January 2012 Intel purchased the InfiniBand product line
from Qlogic for US $125 million in order to fulfill its promise
of developing exascale technology by 2018

4. GOALS OF EXASCALE

5. ARCHITECTURE & MEMORY
CONSUMPTION
(a) Quilt Packaging (b) Thru via chip stack
Reg File
11%
Cache Access
20%
Off-chip
13%
Leakage
28%
On-chip
6%
DRAM Access
1%
FPU
21%

6. ARCHITECTURE CONTD
• Intel neo is a processor that is used in exascale computing.
• Its size in 0.1mm.
• 256 processors are embedded in a circuit around circuit
important components are connected.
• DRAM (Dynamic Random Access Memory) is used for
exascale systems.
• It has a very efficient process.

7. GPU
• A graphics processor unit (GPU), also occasionally called visual
processor unit (VPU), is a specialized electronic circuit designed
to rapidly manipulate and alter memory to accelerate the creation
of images in a frame buffer intended for output to a display
• modern GPUs use most of their transistors to do calculations
related to 3D computer graphics.

8. GETTING TO EXASCALE
• Before 2020, exascale systems will be able to compute a quintillion
operations per second!
• Scientific simulation will continue to push on system requirements:
– To increase the precision of the result
– To get to an answer sooner (e.g., climate modeling, disaster modeling)
• The U.S. will continue to acquire systems of increasing scale
– For the above reasons
– And to maintain competitiveness
• A similar phenomenon in commodity machines
– More, faster, cheaper

9. ADVANTAGES
• GPU interface
• Low time consumption
• It is used in LCF (Leadership Computing Facillity)
• It has high processing speed
• Sustained petascale and Exascale
• High network performance

10. CHALLENGES
• Exascale architectures will be fundamentally different
• Power management becomes fundamental
• Reliability (h/w and s/w) increasingly a concern
• Memory reduction to .01 bytes/flop
• Complex to design
• Need large area

11. SOLUTIONS
• We need to create a new software/architecture to reduce
power consumption
• Express and manage locality and parallelism for billion threads
• Create/support applications that are prepared for new hardware
• If nanotechnology integrated into this we can made this system
to small

12. FUTURE DEVELOPMENT
• In 2018 USA going to launch their first exascale system.
• In Europe Manchester university started their research.
• Japan government launch their exascale system in 2020 with
30 megawatt consumption.
• Intel and DEEP also started their research towards exascale
computing.

13. APPLICATIONS
• Space research
• Biology
• Medicine
• Scientific calculations
• Metallurgy
• Weather
• Material science …etc

14. CONCLUSION
• It’s a emerging technology
• Exascale is a new computing technique is used for processing many
calculations at a second.
• It has a new architectures better than super computing..
• GPU gives very much graphical properties than supercomputing.
• In future the computing are ruled by exascale.