Difference between revisions of "AMGCC18 Program"

From HTC-as-a-Service, KISTI
Jump to: navigation, search
(Workshop Program | Workshop Main Page)
(Workshop Program | Workshop Main Page)
Line 22: Line 22:
 
| 14:10 - 14:30
 
| 14:10 - 14:30
 
| Design and Implementation of Software-based Dynamically Connected Transport
 
| Design and Implementation of Software-based Dynamically Connected Transport
|
 
 
|  
 
|  
  
Line 30: Line 29:
 
| 14:30 - 14:50
 
| 14:30 - 14:50
 
| DymGPU: Dynamic Memory Management for Sharing GPUs in Virtualized Clouds
 
| DymGPU: Dynamic Memory Management for Sharing GPUs in Virtualized Clouds
|
 
 
|  
 
|  
  
Line 36: Line 34:
 
| 14:50 - 15:10
 
| 14:50 - 15:10
 
| Efficient Multi-GPU Memory Management for Deep Learning Acceleration
 
| Efficient Multi-GPU Memory Management for Deep Learning Acceleration
|
 
 
|  
 
|  
  
Line 42: Line 39:
 
| 15:10 - 15:30
 
| 15:10 - 15:30
 
| High Performance Parallel KMP Algorithm on a Heterogeneous Architecture
 
| High Performance Parallel KMP Algorithm on a Heterogeneous Architecture
|
 
 
|  
 
|  
  
Line 52: Line 48:
 
| 16:00 - 16:20
 
| 16:00 - 16:20
 
| Effects of VM Placement Constraints in Heterogeneous Virtual Clusters
 
| Effects of VM Placement Constraints in Heterogeneous Virtual Clusters
|
 
 
|  
 
|  
  
Line 60: Line 55:
 
| 16:20 - 16:40
 
| 16:20 - 16:40
 
| OMBM-ML: An Efficient Memory Bandwidth Management for Ensuring QoS and Improving Server Utilization
 
| OMBM-ML: An Efficient Memory Bandwidth Management for Ensuring QoS and Improving Server Utilization
|
 
 
|  
 
|  
  
Line 66: Line 60:
 
| 16:40 - 17:00
 
| 16:40 - 17:00
 
| Efficient Classification of Application Characteristics by using Hardware Performance Counters with Data Mining
 
| Efficient Classification of Application Characteristics by using Hardware Performance Counters with Data Mining
|
 
 
|  
 
|  
  
Line 72: Line 65:
 
| 17:00 - 17:20
 
| 17:00 - 17:20
 
| Design of Skiplist based key-value store on non-volatile memory
 
| Design of Skiplist based key-value store on non-volatile memory
|
 
 
|  
 
|  
  
Line 78: Line 70:
 
| 17:20 - 17:40
 
| 17:20 - 17:40
 
| Adaptive Sharing of GPUs in Heterogeneous Container Clusters
 
| Adaptive Sharing of GPUs in Heterogeneous Container Clusters
|
 
 
|  
 
|  
  

Revision as of 05:44, 7 August 2018

Workshop Program | Workshop Main Page

  • Date: September 7 (Friday), 2018
  • Time: 02:00 PM - 06:00 PM

This workshop features oral presentations of 9 full papers.

Time Title Institution Session Chair
14:00 - 14:10 Welcome Remarks by AMGCC workshop organizers
14:10 - 14:30 Design and Implementation of Software-based Dynamically Connected Transport TBD
14:30 - 14:50 DymGPU: Dynamic Memory Management for Sharing GPUs in Virtualized Clouds
14:50 - 15:10 Efficient Multi-GPU Memory Management for Deep Learning Acceleration
15:10 - 15:30 High Performance Parallel KMP Algorithm on a Heterogeneous Architecture
15:30 - 16:00 Coffee Break
16:00 - 16:20 Effects of VM Placement Constraints in Heterogeneous Virtual Clusters TBD
16:20 - 16:40 OMBM-ML: An Efficient Memory Bandwidth Management for Ensuring QoS and Improving Server Utilization
16:40 - 17:00 Efficient Classification of Application Characteristics by using Hardware Performance Counters with Data Mining
17:00 - 17:20 Design of Skiplist based key-value store on non-volatile memory
17:20 - 17:40 Adaptive Sharing of GPUs in Heterogeneous Container Clusters
17:40 - 18:00 Closing Remarks