You are on page 1of 4

m

IBM
The High Performance Computing (HPC) group at IBM Research - India is engaged in
designing and analyzing cutting edge parallel programs and improving the performance of
engineering, scientific, and business applications on high performance platforms such as the
IBM Blue Gene supercomputer. The group has following major focus areas:

ám optimization of performance on multi-core processors, large-scale supercomputers,


and clusters
ám parallel scalable algorithms for supercomputers.
ám ptimized Map Reduce for Next Generation Architectures
ám acalable Distributed acheduling for Runtime aystems

Following sections provide more details on the research projects in the HPC area.

Ú  
  
The HPC Challenge (HPCC), a suite of seven benchmarks, is fast evolving as a standard for
evaluating the performance of supercomputers across a spectrum of real-world applications.
The HPC team at IBM Research - India is actively involved in performance optimization and
tuning of the HPC Challenge benchmarks on high-end systems, such as the Blue Gene
supercomputers and P R 7 based systems. The optimizations include designing new
algorithms, data structures and other intricate techniques for distributed memory and
multicore architectures. This group has optimized the aTRAM, RandomAccess, Fast
Fourier Transform and Transpose benchmarks for various systems. The HPC Challenge
RandomAccess benchmark optimized by this team has won the HPCC Class I award at
aupercomputing for all the 5years (2005-2009).


 
The HPC team is also involved in optimizing various scientific applications that require high-
end systems for large scale processing. These applications include molecular dynamics
simulation packages such as VAaP and other simulation programs on the Blue Gene
supercomputers.

     



 

The solution of a system of linear equations generated by discretization of PDs that govern
motion of mobile charge carriers is the most time consuming part of a semiconductor device
simulation. As we go below the 32nm range, the physical models become more complex in
order to address non trivial quantum effects and as a result, the ensuing linear systems
become highly ill- conditioned. Currently, direct solvers are the preferred choice to deal with

m
m

these highly ill-conditioned systems mainly due to their high accuracy. However, in the case
of large 3D simulations, the time for solving these systems is becoming prohibitive. The main
objective of this project is to speed up the total simulation time on architectures such as Blue
Gene, IBM P R, GPU, etc. for both - direct as well as preconditioned iterative solvers.

  !" 


"  
This project aims to design and implement optimizations for Map Reduce runtime system on
next generation architectures such as multi-core clusters and supercomputers. The Map
Reduce paradigm has become very popular for parallel execution of scientific and data
mining applications. Performance optimizations for Map Reduce systems on multi-core
architectures will enable efficient map reduce solutions in terms of cost, power and area.


#  !$ 

 
ptimization of analytics kernels on next generation multi-core architectures has become an
important area of research. As part of this project we are looking at analytics kernels that are
relevant for the amarter Planet domains such as amart Grid, amart ater and so forth. e
aim to design optimized parallel algorithms for graph and data mining algorithms for multi-
core clusters, large aMPs and also supercomputers such as Blue Gene/L.

"
 % Ankur Narang, amruti R. aarangi, Raj Gupta, aouvik Bhattacherjee

Ú
&' & "$ ()* )+


Data Intensive aupercomputing requires processing massive amounts of data using high
number of processors and possibly delivering real-time processing rates. This project studies
the Text Indexing and aearch algorithms on supercomputers such as IBM Blue Gene/L. It
involves design of distributed data-structures and algorithms for massive scale text indexing
and search. e have demonstrated indexing speed of 10GB/s using 4K processors of Blue
Gene/L.

' & 
      
 
acheduling dynamically unfolding computation graphs on distributed multi-core architectures
is an important and challenging problem. This distributed scheduling algorithm needs to
ensure deadlock free execution, should follow locality annotations and also optimize for low
space, time and message complexity. PGAa languages such as X10 expect the programmer to
specify locality annotations to obtain better performance. e study the design of distributed
scheduling algorithms that follow locality while ensuring deadlock free execution under
bounded space per place. This project involves both theoretical analysis of space, time and
message bounds of scheduling algorithms and also implementation of the algorithms and
studying their performance in terms of space-time trade-offs.

m
m

), 

Next Generation Appliances group focuses its activities on architectures and algorithms for
3G and 4G telecommunication networks and private enterprise networks. The group is co-
leading research on the ireless-IT convergence which is emerging as an area of great
opportunity for IBM. ireless-IT convergence explores the opportunities for improving the
performance and impact of future wireless networks through intelligent use of Information
Technology (IT).

The main research threads in the group are:

Ô -*( Ô  %


Increasing number of mobile devices and growing multimedia traffic on wireless networks
introduces severe bottleneck at the wireless link, backhaul and the core network of the
underlying infrastructure. The group is working on cost-effective solutions to reduce this
bottleneck using advanced wireless and IT optimizations. ther research areas include
improving quality-of-service for multimedia streaming over broadband wireless networks,
and optimizations for machine-to-machine communications.

Ô   %
The ireless Network Cloud ( NC) is a novel wireless system architecture that leverages
developments in cloud computing and emerging wireless technologies, such as aoftware
Radio and Remote Radio Header technology. NC allows separation of hardware and
software development for different wireless standards and enables cost savings through
infrastructure sharing between operators. It also opens up many new business models for
network access and service providers. This project is in collaboration with IBM Research -
China to define the NC system requirements and architecture and building a NC proof-
of-concept.

  


Networking requirements for datacenters are changing with cloud computing. As cloud
datacenters scale to tens of thousands of servers, the traditional single rooted, hierarchical and
overprovisioned datacenter network starts becoming a bottleneck from a performance and
manageability perspective. The cloud networking effort at IBM Research - India is exploring
new datacenter network architectures specifically aimed for cloud computing. ne of the
problems that we are currently exploring is cloud traffic modeling and simulation. The
problem is inherently complex due to the numerous layers interacting in the cloud and high
volatility in network traffic. Most of the recent datacenter architectures have been evaluated
on modest real testbeds (upto 100 servers with 10-15 switches). It is hard to predict
performance at 1000s of servers with 100s of switches. Another important aspect of our work
is exploring novel methods of doing networking functions in software.

    

m
m

The world is increasingly becoming a smaller and flatter place. At the same time we are
encountering significant challenges through scarcity of resources and environmental
catastrophes. The amarter Planet Project in IBM Research - India focuses on implementing
fast sensor data mining algorithms on massively multi-threaded hardware for key sectors like
energy and utilities, water, and transportation. Along with these vertical competencies, the
team is also developing assets related to sensing, networking, analytics and high performance
computing.

In transportation, we are interested in traffic estimation and prediction. Traffic measurement


has remained a high priority topic in transportation community. It gained pace with
technologies like inductive-loop in 1960s, video image analysis in 1970s, floating car in
1990s and telco data in 2000s. However, much of the previous work has focused on traffic
dominated by automotives following regular lanes. ur work specifically considers hybrid
traffic seen in developing regions where the traffic pattern is dominated by slow-moving
vehicles like motor-cycles, cycles and pedestrians.

e are not only focusing on the sub-components of amarter Cities, but also bringing them
together in our efforts on amarter Campuses. A key focus here is on visualization (GIa, 3D)
and integration technologies likes business processes and aervice riented Architecture.

In amarter Mobile Finance, the team is looking at the intersection of technologies in


biometrics, mobile, and spoken web for disruptive innovation in mobile payments, mobile
micro-marketing and mobile-enabled microfinance.

You might also like