Skip to main content

suraj kumar

Follower

1

Following

1

Public Views

Jose Luis Cabellos

Nafiseh Abdolahi

Destiny Charlie

Apoorva Dwivedi

Geh Wilson Ejuh

University of Dschang, Cameroon

Interests

Uploads

Papers by suraj kumar

DFT study on the structural, optical and electronic properties of platinum group doped graphene

Materials Today Communications, 2021

This is a PDF file of an article that has undergone enhancements after acceptance, such as the ad... more This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Performance Models for Data Transfers

Proceedings of the 48th International Conference on Parallel Processing, 2019

With increasing complexity of hardwares, systems with different memory nodes are ubiquitous in Hi... more With increasing complexity of hardwares, systems with different memory nodes are ubiquitous in High Performance Computing (HPC). It is paramount to develop strategies to overlap the data transfers between memory nodes with computations in order to exploit the full potential of these systems. In this article, we consider the problem of deciding the order of data transfers between two memory nodes for a set of independent tasks with the objective to minimize the makespan. We prove that with limited memory capacity, obtaining the optimal order of data transfers is a NP-complete problem. We propose several heuristics for this problem and provide details about their favorable situations. We present an analysis of our heuristics on traces, obtained by running 2 molecular chemistry kernels, namely, Hartree-Fock (HF) and Coupled Cluster Single Double (CCSD) on 10 nodes of an HPC system. Our results show that some of our heuristics achieve significant overlap for moderate memory capacities and are very close to the lower bound of makespan.

Analysis of a List Scheduling Algorithm for Task Graphs on Two Types of Resources

2020 IEEE International Parallel and Distributed Processing Symposium (IPDPS), 2020

We consider the problem of scheduling task graphs on two types of unrelated resources, which aris... more We consider the problem of scheduling task graphs on two types of unrelated resources, which arises in the context of task-based runtime systems on modern platforms containing CPUs and GPUs. In this paper, we focus on an algorithm named HeteroPrio, which was originally introduced as an efficient heuristic for a particular application. HeteroPrio is an adaptation of the well known list scheduling algorithm, in which the tasks are picked by the resources in the order of their acceleration factor. This algorithm is augmented with a spoliation mechanism: a task assigned by the list algorithm can later on be reassigned to a different resource if it allows to finish this task earlier.We propose here the first theoretical analysis of the HeteroPrio algorithm in the presence of dependencies. More specifically, if the platform contains m and n processors of each type, we show that the worst-case approximation ratio of HeteroPrio is between $1 + \max \left( {\frac{m}{n},\frac{n}{m}} \right)$ ...

The effectiveness of a robotic tilt table on the muscle strength and quality of life in individuals following stroke: a randomised control trial

International Journal of Therapy and Rehabilitation, 2020

Background/AimsStroke is a common, serious and disabling condition. There is a need for early reh... more Background/AimsStroke is a common, serious and disabling condition. There is a need for early rehabilitation to cut the long-term ill effects of stroke and to maximise functional recovery. This study aimed to assess the effect of Erigo tilt table training on muscle strength and quality of life in individuals who have survived a stroke.MethodsA total of 133 patients were recruited and randomly assigned to receive conventional physiotherapy (group A) or Erigo tilt table training (group B). Both groups underwent an intervention programme lasting 30 days. Quality of life and muscle strength were assessed at baseline (day 0), at the end of the intervention (day 30) and at follow up (day 90).ResultsOverall, 110 patients completed the study: 55 in group A and 55 in group B. Quality of life and muscle strength significantly improved between baseline and follow up with both interventions. Group A's quality of life and upper and lower muscle strength scores increased from 75, 0.57 and 1.0...

Efficacy of Erigo Tilt-Table and Conventional Physiotherapy on Rehabilitation Outcome Within Haemorrhagic and Ischemic Stroke Patients

Journal of Musculoskeletal Research, 2020

Background: Stroke, either haemorrhagic or ischemic in nature, has the ability to culminate in de... more Background: Stroke, either haemorrhagic or ischemic in nature, has the ability to culminate in devastating clinical outcomes thus effective rehabilitation is needed. Objective: To analyze the efficacy of Erigo tilt-table over conventional physiotherapy between ischemic and haemorrhagic stroke patients. Methods: A total of 110 haemorrhagic and ischemic stroke patients (age [Formula: see text] years, [Formula: see text] days after stroke) were assigned randomly into four groups, haemorrhagic conventional physiotherapy group ([Formula: see text]), haemorrhagic Erigo tilt-table group ([Formula: see text]), ischemic conventional physiotherapy group ([Formula: see text]) and ischemic Erigo tilt-table group ([Formula: see text]). The National Institutes of Health Stroke Scale (NIHSS), Mini Mental Scale Examination (MMSE), Modified Ashworth Scale, Quality of Life (QOL) and muscle strength outcomes were assessed at baseline, after the intervention (30th) and on 90th day of follow up. Results...

Comparison between Erigo tilt-table exercise and conventional physiotherapy exercises in acute stroke patients: a randomized trial

Archives of Physiotherapy, 2020

Background Stroke is a common, serious, and disabling health-care problem throughout the world. A... more Background Stroke is a common, serious, and disabling health-care problem throughout the world. Although great advances have been made in acute stroke management, the most of post-stroke care to reduce a patient’s dependency relies on rehabilitation. Purpose To compare the effectiveness of exercises using an Erigo tilt-table and conventional physiotherapy in the rehabilitation of acute stroke patients. Methods A total of 110 acute stroke patients (age 51.08 ± 7.48 years, 8.69 ± 4.62 days after stroke) were assigned randomly into two groups, 55 in each for 30 days of conventional physiotherapy (Group A) or Erigo tilt-table (Group B) rehabilitation. The National Institutes of Health Stroke Scale (NIHSS), Mini-Mental Scale Examination (MMSE), Modified Ashworth Scale were used to measure muscle tone, quality of life (QOL) and muscle strength (MMT), Affected upper (UE) and lower limb (LE) outcomes were assessed at baseline (day 0), after day 30 of the intervention and on 90th day of foll...

Rapidly Progressive Atrial Mass and Cardiac Tamponade

JACC: Case Reports, 2020

Cardiac involvement in multiple myeloma is rare. We report a rapidly progressive atrial mass and ... more Cardiac involvement in multiple myeloma is rare. We report a rapidly progressive atrial mass and plasma cell-infiltrated pericardial effusion with tamponade. Bone marrow biopsy and plasma immunoelectrophoresis confirmed multiple myeloma (Revised International Staging System Stage III). The patient died within 18 weeks of presentation, suggesting aggressive disease with poor prognosis.

Fast approximation algorithms for task-based runtime systems

Concurrency and Computation: Practice and Experience, 2018

In High Performance Computing, heterogeneity is now the norm with specialized accelerators like G... more In High Performance Computing, heterogeneity is now the norm with specialized accelerators like GPUs providing efficient computational power. Resulting complexity led to the development of task-based runtime systems, where complex computations are described as task graphs, and scheduling decisions are made at runtime to perform load balancing between all resources of the platforms. Developing good scheduling strategies, even at the scale of a single node, and analyzing them both theoretically and in practice is expected to have a very high impact on the performance of current HPC systems. The special case of two kinds of resources, typically CPUs and GPUs is already of great practical interest. The scheduling policy Hetero-Prio has been proposed in the context of fast multipole computations (FMM), and has been extended to general task graphs with very promising results. In this paper, we provide a theoretical study of the performance of HeteroPrio, by proving approximation bounds compared to the optimal schedule, both in the case of independent tasks and in the case of general task graphs. Interestingly, our results establish that spoliation (a technique that enables resources to restart uncompleted tasks on another resource) is enough to prove bounded approximation ratios for a list scheduling algorithm on two unrelated resources, which is known to be impossible otherwise. This result holds true both for independent and dependent tasks graphs. Additionally, we provide an experimental evaluation of HeteroPrio on real task graphs from dense linear algebra computation, that establishes its strong performance in practice.

Are Static Schedules so Bad? A Case Study on Cholesky Factorization

2016 IEEE International Parallel and Distributed Processing Symposium (IPDPS), 2016

Our goal is to provide an analysis and comparison of static and dynamic strategies for task graph... more Our goal is to provide an analysis and comparison of static and dynamic strategies for task graph scheduling on platforms consisting of heterogeneous and unrelated resources, such as GPUs and CPUs. Static scheduling strategies, that have been used for years, suffer several weaknesses. First, it is well known that underlying optimization problems are NP-Complete, what limits the capability of finding optimal solutions to small cases. Second, parallelism inside processing nodes makes it difficult to precisely predict the performance of both communications and computations, due to shared resources and co-scheduling effects. Recently, to cope with this limitations, many dynamic task-graph based runtime schedulers (StarPU, StarSs, QUARK, PaRSEC) have been proposed. Dynamic schedulers base their allocation and scheduling decisions on the one side on dynamic information such as the set of available tasks, the location of data and the state of the resources and on the other hand on static information such as task priorities computed from the whole task graph. Our analysis is deep but we concentrate on a single kernel, namely Cholesky factorization of dense matrices on platforms consisting of GPUs and CPUs. This application encompasses many important characteristics in our context. Indeed, it involves 4 different kernels (POTRF, TRSM, SYRK and GEMM) whose acceleration ratios on GPUs are strongly different (from 2.3 for POTRF to 29 for GEMM) and it consists in a phase where the number of available tasks if large, where the careful use of resources is critical, and in a phase with few tasks available, where the choice of the task to be executed is crucial. In this paper, we analyze the performance of static and dynamic strategies and we propose a set of intermediate strategies, by adding more static (resp. dynamic) features into dynamic (resp. static) strategies. Our conclusions are somehow unexpected in the sense that we prove that static-based strategies are very efficient, even in a context where performance estimations are not very good.

Chapter-14 Microbiology

A Complete Textbook for GNM Internship, 2015

Bridging the Gap between Performance and Bounds of Cholesky Factorization on Heterogeneous Platforms

2015 IEEE International Parallel and Distributed Processing Symposium Workshop, 2015

We consider the problem of allocating and scheduling dense linear application on fully heterogene... more We consider the problem of allocating and scheduling dense linear application on fully heterogeneous platforms made of CPUs and GPUs. More specifically, we focus on the Cholesky factorization since it exhibits the main features of such problems. Indeed, the relative performance of CPU and GPU highly depends on the sub-routine: GPUs are for instance much more efficient to process regular kernels such as matrix-matrix multiplications rather than more irregular kernels such as matrix factorization. In this context, one solution consists in relying on dynamic scheduling and resource allocation mechanisms such as the ones provided by PaRSEC or StarPU. In this paper we analyze the performance of dynamic schedulers based on both actual executions and simulations, and we investigate how adding static rules based on an offline analysis of the problem to their decision process can indeed improve their performance, up to reaching some improved theoretical performance bounds which we introduce.

Evaluation of Xpert MTB/RIF assay performance in diagnosing extrapulmonary tuberculosis among adults in a tertiary care centre in India

The European respiratory journal, 2014

A DSL customer-premise equipment modem SoC with extended reach/rate for broadband bridging and routing

2004 IEEE International Solid-State Circuits Conference (IEEE Cat. No.04CH37519)

This single chip DSL CPE Modem supports multifarious stan-dards like G.dmt.992 Annex A, B, C, G.d... more

DFT study on the structural, optical and electronic properties of platinum group doped graphene

Materials Today Communications, 2021

This is a PDF file of an article that has undergone enhancements after acceptance, such as the ad... more This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Performance Models for Data Transfers

Proceedings of the 48th International Conference on Parallel Processing, 2019

With increasing complexity of hardwares, systems with different memory nodes are ubiquitous in Hi... more With increasing complexity of hardwares, systems with different memory nodes are ubiquitous in High Performance Computing (HPC). It is paramount to develop strategies to overlap the data transfers between memory nodes with computations in order to exploit the full potential of these systems. In this article, we consider the problem of deciding the order of data transfers between two memory nodes for a set of independent tasks with the objective to minimize the makespan. We prove that with limited memory capacity, obtaining the optimal order of data transfers is a NP-complete problem. We propose several heuristics for this problem and provide details about their favorable situations. We present an analysis of our heuristics on traces, obtained by running 2 molecular chemistry kernels, namely, Hartree-Fock (HF) and Coupled Cluster Single Double (CCSD) on 10 nodes of an HPC system. Our results show that some of our heuristics achieve significant overlap for moderate memory capacities and are very close to the lower bound of makespan.

Analysis of a List Scheduling Algorithm for Task Graphs on Two Types of Resources

2020 IEEE International Parallel and Distributed Processing Symposium (IPDPS), 2020

We consider the problem of scheduling task graphs on two types of unrelated resources, which aris... more We consider the problem of scheduling task graphs on two types of unrelated resources, which arises in the context of task-based runtime systems on modern platforms containing CPUs and GPUs. In this paper, we focus on an algorithm named HeteroPrio, which was originally introduced as an efficient heuristic for a particular application. HeteroPrio is an adaptation of the well known list scheduling algorithm, in which the tasks are picked by the resources in the order of their acceleration factor. This algorithm is augmented with a spoliation mechanism: a task assigned by the list algorithm can later on be reassigned to a different resource if it allows to finish this task earlier.We propose here the first theoretical analysis of the HeteroPrio algorithm in the presence of dependencies. More specifically, if the platform contains m and n processors of each type, we show that the worst-case approximation ratio of HeteroPrio is between $1 + \max \left( {\frac{m}{n},\frac{n}{m}} \right)$ ...

The effectiveness of a robotic tilt table on the muscle strength and quality of life in individuals following stroke: a randomised control trial

International Journal of Therapy and Rehabilitation, 2020

Background/AimsStroke is a common, serious and disabling condition. There is a need for early reh... more Background/AimsStroke is a common, serious and disabling condition. There is a need for early rehabilitation to cut the long-term ill effects of stroke and to maximise functional recovery. This study aimed to assess the effect of Erigo tilt table training on muscle strength and quality of life in individuals who have survived a stroke.MethodsA total of 133 patients were recruited and randomly assigned to receive conventional physiotherapy (group A) or Erigo tilt table training (group B). Both groups underwent an intervention programme lasting 30 days. Quality of life and muscle strength were assessed at baseline (day 0), at the end of the intervention (day 30) and at follow up (day 90).ResultsOverall, 110 patients completed the study: 55 in group A and 55 in group B. Quality of life and muscle strength significantly improved between baseline and follow up with both interventions. Group A's quality of life and upper and lower muscle strength scores increased from 75, 0.57 and 1.0...

Efficacy of Erigo Tilt-Table and Conventional Physiotherapy on Rehabilitation Outcome Within Haemorrhagic and Ischemic Stroke Patients

Journal of Musculoskeletal Research, 2020

Background: Stroke, either haemorrhagic or ischemic in nature, has the ability to culminate in de... more Background: Stroke, either haemorrhagic or ischemic in nature, has the ability to culminate in devastating clinical outcomes thus effective rehabilitation is needed. Objective: To analyze the efficacy of Erigo tilt-table over conventional physiotherapy between ischemic and haemorrhagic stroke patients. Methods: A total of 110 haemorrhagic and ischemic stroke patients (age [Formula: see text] years, [Formula: see text] days after stroke) were assigned randomly into four groups, haemorrhagic conventional physiotherapy group ([Formula: see text]), haemorrhagic Erigo tilt-table group ([Formula: see text]), ischemic conventional physiotherapy group ([Formula: see text]) and ischemic Erigo tilt-table group ([Formula: see text]). The National Institutes of Health Stroke Scale (NIHSS), Mini Mental Scale Examination (MMSE), Modified Ashworth Scale, Quality of Life (QOL) and muscle strength outcomes were assessed at baseline, after the intervention (30th) and on 90th day of follow up. Results...

Comparison between Erigo tilt-table exercise and conventional physiotherapy exercises in acute stroke patients: a randomized trial

Archives of Physiotherapy, 2020

Background Stroke is a common, serious, and disabling health-care problem throughout the world. A... more Background Stroke is a common, serious, and disabling health-care problem throughout the world. Although great advances have been made in acute stroke management, the most of post-stroke care to reduce a patient’s dependency relies on rehabilitation. Purpose To compare the effectiveness of exercises using an Erigo tilt-table and conventional physiotherapy in the rehabilitation of acute stroke patients. Methods A total of 110 acute stroke patients (age 51.08 ± 7.48 years, 8.69 ± 4.62 days after stroke) were assigned randomly into two groups, 55 in each for 30 days of conventional physiotherapy (Group A) or Erigo tilt-table (Group B) rehabilitation. The National Institutes of Health Stroke Scale (NIHSS), Mini-Mental Scale Examination (MMSE), Modified Ashworth Scale were used to measure muscle tone, quality of life (QOL) and muscle strength (MMT), Affected upper (UE) and lower limb (LE) outcomes were assessed at baseline (day 0), after day 30 of the intervention and on 90th day of foll...

Rapidly Progressive Atrial Mass and Cardiac Tamponade

JACC: Case Reports, 2020

Cardiac involvement in multiple myeloma is rare. We report a rapidly progressive atrial mass and ... more Cardiac involvement in multiple myeloma is rare. We report a rapidly progressive atrial mass and plasma cell-infiltrated pericardial effusion with tamponade. Bone marrow biopsy and plasma immunoelectrophoresis confirmed multiple myeloma (Revised International Staging System Stage III). The patient died within 18 weeks of presentation, suggesting aggressive disease with poor prognosis.

Fast approximation algorithms for task-based runtime systems

Concurrency and Computation: Practice and Experience, 2018

In High Performance Computing, heterogeneity is now the norm with specialized accelerators like G... more In High Performance Computing, heterogeneity is now the norm with specialized accelerators like GPUs providing efficient computational power. Resulting complexity led to the development of task-based runtime systems, where complex computations are described as task graphs, and scheduling decisions are made at runtime to perform load balancing between all resources of the platforms. Developing good scheduling strategies, even at the scale of a single node, and analyzing them both theoretically and in practice is expected to have a very high impact on the performance of current HPC systems. The special case of two kinds of resources, typically CPUs and GPUs is already of great practical interest. The scheduling policy Hetero-Prio has been proposed in the context of fast multipole computations (FMM), and has been extended to general task graphs with very promising results. In this paper, we provide a theoretical study of the performance of HeteroPrio, by proving approximation bounds compared to the optimal schedule, both in the case of independent tasks and in the case of general task graphs. Interestingly, our results establish that spoliation (a technique that enables resources to restart uncompleted tasks on another resource) is enough to prove bounded approximation ratios for a list scheduling algorithm on two unrelated resources, which is known to be impossible otherwise. This result holds true both for independent and dependent tasks graphs. Additionally, we provide an experimental evaluation of HeteroPrio on real task graphs from dense linear algebra computation, that establishes its strong performance in practice.

Are Static Schedules so Bad? A Case Study on Cholesky Factorization

2016 IEEE International Parallel and Distributed Processing Symposium (IPDPS), 2016

Our goal is to provide an analysis and comparison of static and dynamic strategies for task graph... more Our goal is to provide an analysis and comparison of static and dynamic strategies for task graph scheduling on platforms consisting of heterogeneous and unrelated resources, such as GPUs and CPUs. Static scheduling strategies, that have been used for years, suffer several weaknesses. First, it is well known that underlying optimization problems are NP-Complete, what limits the capability of finding optimal solutions to small cases. Second, parallelism inside processing nodes makes it difficult to precisely predict the performance of both communications and computations, due to shared resources and co-scheduling effects. Recently, to cope with this limitations, many dynamic task-graph based runtime schedulers (StarPU, StarSs, QUARK, PaRSEC) have been proposed. Dynamic schedulers base their allocation and scheduling decisions on the one side on dynamic information such as the set of available tasks, the location of data and the state of the resources and on the other hand on static information such as task priorities computed from the whole task graph. Our analysis is deep but we concentrate on a single kernel, namely Cholesky factorization of dense matrices on platforms consisting of GPUs and CPUs. This application encompasses many important characteristics in our context. Indeed, it involves 4 different kernels (POTRF, TRSM, SYRK and GEMM) whose acceleration ratios on GPUs are strongly different (from 2.3 for POTRF to 29 for GEMM) and it consists in a phase where the number of available tasks if large, where the careful use of resources is critical, and in a phase with few tasks available, where the choice of the task to be executed is crucial. In this paper, we analyze the performance of static and dynamic strategies and we propose a set of intermediate strategies, by adding more static (resp. dynamic) features into dynamic (resp. static) strategies. Our conclusions are somehow unexpected in the sense that we prove that static-based strategies are very efficient, even in a context where performance estimations are not very good.

Chapter-14 Microbiology

A Complete Textbook for GNM Internship, 2015

Bridging the Gap between Performance and Bounds of Cholesky Factorization on Heterogeneous Platforms

2015 IEEE International Parallel and Distributed Processing Symposium Workshop, 2015

We consider the problem of allocating and scheduling dense linear application on fully heterogene... more We consider the problem of allocating and scheduling dense linear application on fully heterogeneous platforms made of CPUs and GPUs. More specifically, we focus on the Cholesky factorization since it exhibits the main features of such problems. Indeed, the relative performance of CPU and GPU highly depends on the sub-routine: GPUs are for instance much more efficient to process regular kernels such as matrix-matrix multiplications rather than more irregular kernels such as matrix factorization. In this context, one solution consists in relying on dynamic scheduling and resource allocation mechanisms such as the ones provided by PaRSEC or StarPU. In this paper we analyze the performance of dynamic schedulers based on both actual executions and simulations, and we investigate how adding static rules based on an offline analysis of the problem to their decision process can indeed improve their performance, up to reaching some improved theoretical performance bounds which we introduce.

Evaluation of Xpert MTB/RIF assay performance in diagnosing extrapulmonary tuberculosis among adults in a tertiary care centre in India

The European respiratory journal, 2014

A DSL customer-premise equipment modem SoC with extended reach/rate for broadband bridging and routing

2004 IEEE International Solid-State Circuits Conference (IEEE Cat. No.04CH37519)

This single chip DSL CPE Modem supports multifarious stan-dards like G.dmt.992 Annex A, B, C, G.d... more