Cid

1
Cid

• CS498LVK
• 4 April 2006
• Aaron Becker
• Abhinav Bhatele
• Isaac Dooley

2
Overview

• Model MIMD threads with lock-protected shared
  data
• Intended to be similar to SMP threaded programs
• Preprocessor for standard C compiler

3
Cid is C With

• Global pointers
• global int n
• cid_get(n, CID_READ)
• cid_rel(n)
• Threads spawned on remote processors
• cid_fork(jv )
• do_work()
• cid_jwait(jv)
• More stuff well talk about later

4

• struct node_s
• int info
• global struct node_s left
• global struct node_s right
• node
• cid_forkable
• global node build_tree(int d)
• node nodep
• cid_jvar jv CID_JVAR_INITIAL
• if (d 0)
• return CID_NULL
• else
• nodep (node) malloc(sizeof(node))
• cid_fork(jv )
• nodep-gtleft build_tree(d-1)
• cid_fork(jv )
• nodep-gtright build_tree(d-1)

5

• cid_forkable
• int sum_tree(global node nodep)
• int i, s1, s2
• cid_jvar jv CID_JVAR_INITIAL
• if (nodep CID_NULL)
• return 0
• else
• cid_get(nodep, CID_READ)
• cid_fork(jv )
• s1 sum_tree(nodep-gtleft)
• cid_fork(jv )
• s2 sum_tree(nodep-gtright)
• i nodep-gtinfo
• cid_rel(nodep)
• cid_jwait(jv)
• return i s1 s2

Whats wrong with this approach?
6

• cid_forkable
• int sum_tree(global node nodep)
• int i, s1, s2
• cid_jvar jv CID_JVAR_INITIAL
• if (nodep CID_NULL)
• return 0
• else
• cid_get(nodep, CID_READ)
• cid_fork(jv cid_to_pe(nodep-gtleft))
• s1 sum_tree(nodep-gtleft)
• cid_fork(jv cid_to_pe(nodep-gtright))
• s2 sum_tree(nodep-gtright)
• i nodep-gtinfo
• cid_rel(nodep)
• cid_jwait(jv)
• return i s1 s2

7

• cid_forkable
• int sum_graph(global node nodep)
• int i, s1, s2
• cid_jvar jv CID_JVAR_INITIAL
• if (nodep CID_NULL)
• return 0
• else
• cid_get(nodep, CID_WRITE)
• if (nodep-gtmark TRUE)
• cid_rel(nodep)
• return 0
• else
• nodep-gtmark TRUE
• cid_fork(jv cid_to_pe(nodep-gtleft))
• s1 sum_graph(nodep-gtleft)
• cid_fork(jv cid_to_pe(nodep-gtright))
• s2 sum_graph(nodep-gtright)
• i nodep-gtinfo

8
Automatic Load Balancing

• On fork, if destination PE not specified, runtime
  attempts to choose underutilized processor
• Work-stealing scheduler attempts to balance load
  dynamically

9
Accumulators

• for (j0 jltN j)
• cid_fork()
• resultsj f()
• s 0
• for (j0 jltN j)
• s resultsj

10
Accumulators

• s 0
• for (j0 jltN j)
• cid_fork()
• s f()

11
Distributed Arrays

• Similar to HPF
• cid_alloc_2d(jv, gp, NI, NJ,
• sizeof_elem, distrib,
• block_factor)

block, cyclic, etc.
size of dist. unit