Pratap Pattnaik

FlexRAM: Toward an advanced Intelligent Memory system

Major advances in Merged Logic DRAM (MLD) technology coupled with the popularization of memory-intensive applications provide fertile ground for architectures based on Intelligent Memory (IRAM) or Processors-in-Memory (PIM). The contribution of this paper is to explore one way to use the current state-of-the-art MLD technology for general-purpose computers. To satisfy requirements of general purpose and low programming cost, we place the PIM chips in the memory system and let them default to plain DRAM if the application is not enabled for intelligent memory. Since wide usability is crucial, we identify and analyze a range of real applications for PIM. Based on the requirements of these applications and current technological constraints, we design a PIM chip and a PIM-based memory system. We call the chip FlexRAM. We describe FlexRAMs design and floorplan, and the resulting memory system. Evaluation of the system through simulations shows that 4 FlexRAM chips often allow a workstation to run 25-40 times faster.

Modeling of Workload in MPPs

In this paper we have characterized the inter-arrival time and service time distributions for jobs at a large MPP supercomputing center. Our findings show that the distributions are dispersive and complex enough that they require Hyper Erlang distributions to capture the first three moments of the observed workload. We also present the parameters from the characterization so that they can be easily used for both theoretical studies and the simulations of various scheduling algorithms.

The M2 channel of influenza A virus: a molecular dynamics study

Molecular dynamics simulations have been performed on a tetramer of the 25‐residue (SSDPLVVAASIIGILHLILWILDRL) synthetic peptide [1] which contains the transmembrane domain of the influenza A virus M2 coat protein. The peptide bundle was initially assembled as a parallel α‐helix bundle in the octane portion of a phase separated water/octane system, which provided a membrane‐mimetic environment. A 4‐ns dynamics trajectory identified a left‐handed coiled coil state of the neutral bundle, with a water filled funnel‐like structural motif at the N‐terminus involving the long hydrophobic sequence. The neck of the funnel begins at V27 and terminates at H37, which blocks the channel. The C‐terminus is held together by inter‐helix hydrogen bonds and contains water below H37. Solvation of the S23 and D24 residues, located at the rim of the funnel, appears to be important for stability of the structure. The calculated average tilt of the helices in the neutral bundle is 27±5°, which agrees well with recent NMR data.

An Evaluation of Parallel Job Scheduling for ASCI Blue-Pacific

In this paper we analyze the behavior of a gang-scheduling system that we are developing for the ASCI Blue-Pacific machines. Starting with a real workload obtained from job logs of one of the ASCI machines, we generate a statistical model of this workload using Hyper Erlang distributions. We then vary the parameters of those distributions to generate various workloads, representative of different operating points of the machine. Through simulation we obtain performance characteristics for three different scheduling strategies: (i) first-come first-serve, (ii) gang-scheduling, and (iii) backfilling. Our results show that both backfilling and gang-scheduling with moderate multiprogramming levels are much more effective than simple first-come first-serve scheduling. In addition, we show that gang-scheduling can display better performance characteristics than backfilling, particularly for large production jobs.

Gang scheduling for highly efficient, distributed multiprocessor systems

We have implemented a job scheduling system for workstation clusters and massively parallel systems with highly efficient message passing interconnects that supports space and time sharing through multiuser gang scheduling of parallel jobs. The system is available on the IBM-SP-2 cluster. It is highly modular, scalable and can easily be adapted to a variety of other MPP systems. The system supports various scheduling policies. We architect the system so that the time-sharing of processors avoids any significant serialization and extra resource consumption, but preserves the reliability and the efficiency of the high performance communication subsystem that characterizes a dedicated non time shared systems.

Constant pressure and temperature molecular-dynamics simulation of the hydrated diphytanolphosphatidylcholine lipid bilayer

Diphytanolphosphatidylcholine (DPhPC) is a lipid widely used in the study of membrane channel activity. Herein we report the results of a constant temperature (T=25 °C) and constant pressure (p=1 atm) molecular-dynamics (MD) simulation of a hydrated liquid crystal phase DPhPC bilayer. The simulated system consisted of a periodically replicated cell containing 64 lipid and 1792 water molecules. The system was monitored during a trajectory spanning more than one nanosecond. The resulting unconstrained area density per lipid agreed quantitatively with experimental data. The calculated bilayer profile and acyl chain order parameters also compared favorably with x-ray scattering and nuclear magnetic resonance (NMR) data.

An Efficient Implementation of MPI

MPI-F, a prototype high-performance implementation of MPI on the IBM SPI is described and MPI-F communication performance is presented.

Two possible conducting states of the influenza A virus M2 ion channel

Molecular dynamics simulations have been performed on protonated four‐helix bundles based on the 25‐residue Duff–Ashley transmembrane sequence of the M2 channel of the influenza A virus. Well‐equilibrated tetrameric channels, with one, two and four of the H37 residues protonated, were investigated. The protonated peptide bundles were immersed in the octane portion of a phase‐separated water/octane system, which provided a membrane‐mimetic environment. The simulations suggest that there could be two conducting states of the M2 channel corresponding to tetramers containing one or two protonated histidines. The more open structure of the doubly protonated state suggests it would have the higher conductance.

Modeling and simulating flash based solid-state disks for operating systems

Solid-State Disks (SSDs) made out of Flash devices have gained a lot of prominence in recent years due to their increasing performance and endurance. A number of mechanisms are being proposed to improve the performance and reliability of these devices from technological and operating system perspectives, to integrate them into personal computers and enterprise systems. Most of such proposals are being implemented and evaluated directly on top of these SSDs and require sophisticated framework and infrastructure for thorough performance evaluation. On the other hand, to our knowledge, very little has been done on modeling Flash devices and building efficient Flash simulators that can be used to simulate SSDs. Such models and simulators can give insights to make design decisions, save a lot of cumbersome work for setup and implementation, save hardware costs and allow researchers to focus on the real methods that are being proposed. This paper presents a linear model for NAND-based Flash devices based on the internal architecture of these devices. Parameters of the model are presented along with micro-benchmarks that can be used to extract these parameters. The model is validated on the STEC Zeus Flash SSD and extracted parameters are used to build a Flash simulator as a kernel extension in the AIX operating system. A key feature of the simulator is that it simulates I/O requests by maintaining minimal state information and is independent of the internal organization of a Flash SSD. The simulator is validated using commercial and raw-IO applications through experimentation on the simulator and real Flash disks.

A Gang Scheduling Design for Multiprogrammed Parallel Computing Environments

Gang scheduling is a resource management scheme for parallel and distributed systems that combines time-sharing and space-sharing to ensure high overall system throughput and short response times for interactive tasks. We recently participated in the design and implementation of a flexible gang scheduling scheme on an IBM SP2 parallel system and a cluster of IBM RS/6000 workstations. In this paper, we present our gang scheduling system and some results of a mathematical model for our system. Using this model, we can obtain exact solutions for measures of system performance as a function of scheduling policy parameters, and thus determine optimal values for several system and policy variables such as the amount of time allocated to the time-slice of each task.