Date post: | 02-Jan-2016 |
Category: |
Documents |
Upload: | yetta-owen |
View: | 43 times |
Download: | 2 times |
ICCS 2004 Gava 1
Frédéric Gava
Design ofDepartmental
Metacomputing ML
ICCS 2004 Gava 2
Goal
User
Computing Unit
Storage
Computing Unit
ICCS 2004 Gava 3
Bulk Synchronous Parallelism +
Functional Programming = BSML
Advantages of the BSP model:1. Portability2. Scalability, deadlock free
3. Simple cost model Performance prediction
Advantages of functional programming:
1. High level features (higher order functions, pattern-matching, concrete types, etc…)
2. Savety of the environment3. Programs Proofs
4
The BSP model
T(s) = (max0i<p wi) + hg + L
0 1 2 3 p-1Proc.
5
The BSML language
Library for the « Objective Caml » language (called BSMLlib)
4 Operations on a parallel data structure called parallel vector
Operations to access to the BSP parameters
mkpar : create a parallel vector
apply : parallel point-wise application
put : send values
at : projection of a value
Confluent language : deterministic algorithms
ICCS 2004 Gava 6
Problems with BSP
No hierarchical network capacities
Global synchronization is too expensive
No heterogenelity
7
The MP model
Local computing
Communication
No super-steps but m-steps
Proc. 0 1 2 3
MP = messages passing
ICCS 2004 Gava 8
New model: DMM We have :
• P BSP units• L, latency of the global network• G, time to send one word
List of nb of processes :
List of times for a barrier :
List of times for a 1-relation :
We have d-steps of super-steps
ICCS 2004 Gava 9
New communicationG et L
g et l
BSP Unit jBSP Unit i
j jgi et li
Intranet
ICCS 2004 Gava 10
New communicationUnité jUnit i
gi et li
New language: DMML Library for the « Objective Caml » language
Operations to access to the DMM parameters
Operations on departmental vectors
mkdpt : creation of a departmental vector
applydpt : departmental point-wise application
get : received values
ICCS 2004 Gava 12
Conclusion BSML = BSP + ML
New model for departmental meta-computing
New language called DMML
ICCS 2004 Gava 13
Future Work Polymorphic type system for the language
Implementation of « big » applications
Static Dependent Costs for Estimating Execution Time
ICCS 2004 Gava 14