Presentation is loading. Please wait.

Presentation is loading. Please wait.

Remote Function Invocation

Similar presentations


Presentation on theme: "Remote Function Invocation"— Presentation transcript:

1 Remote Function Invocation
Running a Single Program On Multiple Computers

2 Thesis Single Program Single Computer Serial Processing

3 Antithesis Multiple Programs Multiple Computers
Cooperate on Single Problem via Messages

4 Synthesis Single Program Multiple Computers

5 Raises Questions Existing mechanisms? Message-passing necessary?
Comparison of Metrics? Anything we can exploit? Use?

6 Objectives Develop the mechanism Benchmark RFI Demonstrate:
Running existing program on cluster Exploitation of new opportunities presented by RFI

7 Programs Native Code Virtual Platform RFI exists

8 Developing Rexel Remote Execution Environment for Linux

9 Node Setup Simple Comms Store libfile.so Register function
Call function

10 Function Calling Few major problems Trade-offs acceptable
Uses dlopen() and dlsym() Uses Libffi for calling convention

11 Usage Usage (mostly) automatic Needs runtime linkage librexelmaster.so
Configured exclusively from config files

12 Test Setup Benchmarked against MPI Setup carefully controlled
System representative of cloud providers Problem-space exhaustion

13 Speed (1/2)

14 Speed (2/2)

15 Speed Overall Predictability

16 Consistency (1/2)

17 Consistency (2/2)

18 Summary Rexel MPI Quicker at lots of smaller problems
Slower when number of repeated calls increase Predictable and consistent throughout Slower when problem is divided into tiny chunks Faster as problem-size of each chunk grows Unpredictable with tested criteria

19 Future? Rexel MPI Performance improved towards the end
Consistency improved towards the end

20 Running SQLite (1/2) SQLite: Small standalone database
Has c/line interface Stores all data in a single file

21 SQLite (2/2) Run in LD_PRELOAD environment Unaware of Rexel
Mixed results Successfully wrote files Nodes could not return errors

22 Partial Shared-Memory
Expressions represented as trees Leaf nodes are operands Branch nodes are operators

23 Solving Expressions Solved recursively Independently

24 Concurrent Solving

25 Research Answers Mechanisms exist
Replacement possible for both shared-mem and MPI Large effects on software design Multiple opportunities to exploit

26 Future Opportunities Can have an impact on program design
Quorum Solving Fine-grained Unit-testing More???


Download ppt "Remote Function Invocation"

Similar presentations


Ads by Google