1. Atlas: An Infrastructure for Global Computing

2. People Eric Baldeschwieler (UC Berkeley) Bobby Blumofe (UT Austin)
Eric Brewer (UC Berkeley)

3. Outline Introduction Programming model Architecture Examples
Discussion
Limitations & Conclusion

4. Introduction Properties of a Internet computing infrastructure
Scalability: to 106 nodes
Heterogeneity: of machines & OSs
Fault tolerance: completion probability comparable to sequential program
Adaptive parallelism: dynamic set of resources

5. Properties ... Safety: Hosts must be secure
Anonymity: Secure privacy of client: data & program
Hierarchy: Locality of communication (local bandwidth typically is higher)
Ease of use: Minimize “costs” of participating.
Reasonable performance: Low overhead  Benefit from a small set of machines.

6. Introduction ... Atlas combines mechanisms from: Java “ensures”: Cilk
with new mechanisms.
Java “ensures”:
heterogeneity
safety

7. Introduction ...
Atlas:
extends Cilk’s work-stealing scheduler to a hierarchical Internet setting
uses Cilk-NOW’s mechanisms for:
adaptive parallelism
fault tolerance

8. Programming Model Applications are written in Java
When a native library is used, heterogeneity is limited to platforms that support it.
Programming model is:
a Java-based implementation of Cilk:
Non-blocking, explicit continuation passing threads
a Unix-like URL-based file system & local caching with coherence.

9. Native libraries (C or C++)
Architecture
Basic architecture
Compute Server
Client
Manager
Application (Java)
Runtime library
Java interpreter
Native libraries (C or C++)
Compute
Server
Compute
Server
Compute
Server

10. Architecture ... Client is a Java application
connects to compute servers on machines other than its manager’s.
Idle servers steal work from busy ones.

11. Architecture Compute server:
relinquishes control when there is non-Atlas work (a screensaver?)
Runs as a daemon:
working
pings manager & siblings for work to steal

12. Architecture: Porting Atlas
A Java runtime system
Port:
natively written URL-based file system
some support routines.

13. Hierarchical Work Stealing
Manager
Manager
Manager
Manager
Manager
Compute
Server
Compute
Server
Compute
Server

14. Hierarchical Work Stealing ...
Manager keeps track of when its subtree is idle
If manager’s subtree is idle,
manager steals work from its siblings
If a subtree has “too much” work,
it “allows” work stealing from above
What is definition & implementation of “too much”?

15. Hierarchical Work Stealing
The authors claim that proven properties of Cilk hold in this hierarchical setting.
Goals:
Localize communication
Sub-trees map to domain hierarchy
Administrators can control thread migration:
Outflow: Privacy
Inflow: Host security

16. Examples Fib: fine grained threads POV-Ray: coarse grained threads
Base 1 Node 3 Nodes 8 Nodes
Fib (24) (2.0) 31 (2.6)
POV-Ray (7.8)
Numbers in ( ) are speedups over 1-node case.

17. Examples ... POV-Ray is not written in Java
Partitioning is done in Java
8 nodes: only 2% overhead.
What about larger P?

18. Discussion Scalable: Yes.
Heterogeneity: Incomplete until divorces itself from all native libraries.
Safety:
Java: OK.
Native libraries: ?

19. Discussion ...
Fault tolerance: A timed out thread is recomputed from a checkpoint maintained by subtree (manager?)
What is affect on performance of checkpointing?
Subtree rooted at a thread is its subcomputation.

20. Fault Tolerance ... Subcomputations are transactions:
Authors claim: side effects can be undone
How does this relate to hierarchical work stealing?

21. Discussion ...
Anonymity: A host executing a stolen subtree cannot determine client.
Managers are assumed to be trustworthy
Hierarchy: Yes, via manager hierarchy.
Ease of use: Interface incomplete.
clients submit jobs via a special “shell”

22. Discussion ... Adaptive parallelism:
“Owner” (?) of compute server sets a policy that defines when server is idle.
How?
When compute server becomes unavailable for Atlas work, all its sub-computations are moved to another computer server.

23. Adaptive Parallelism ...
Moving a subcomputation requires updating information linking subcomputation to its:
parent
children
How long does it take to retreat?
Is sub-computation restarted? From checkpoint?

24. Limitations
Atlas inherits tree-structured program limitation from Cilk.
But this is still a rich set!
Generalizing to non-tree-structured programs seems hard.
No shared variables among threads.
Global file system is read-only.

25. Conclusion Jicos design goals = those for Atlas.
Use JXTA to give Jicos a “file system”
Then, Jicos becomes Atlas’s heir.