Presentation on theme: "Scalable Reader-Writer Synchronization for Shared- Memory Multiprocessors Mellor-Crummey and Scott Presented by Robert T. Bauer."— Presentation transcript:
Scalable Reader-Writer Synchronization for Shared- Memory Multiprocessors Mellor-Crummey and Scott Presented by Robert T. Bauer
Problem Efficient SMMP Reader/Writer Synchronization
Basics Readers can “share” a data structure Writers need exclusive access –Write appears to be atomic Issues: –Fairness: Fair every “process” eventually runs –Preference: Reader preference Writer can starve Writer preference Reader can starve
Organization Algorithm 1 – simple mutual exclusion Algorithm 2 – RW with reader preference Algorithm 3 – A fair lock Algoirthm 4 – local only spinning (Fair) Algorithm 5– local only reader preference Algorithm 6 – local only writer preference Conclusions Paper’s Contrib
Algorithm I – just a spin lock Idea is that processors spin on their own lock record Lock records form a linked list When a lock is released, the “next” processor waiting on the lock is signaled by passing the lock By using “compare-swap” when releasing, the algorithm guarantees FIFO Spinning is “local” by design
Algorithm 1 Acquire Lock pred := fetch_and_store(L, I) pred /= null I->locked := true pred next := I repeat while I locked Release Lock I next == null compare_and_swap(L,I,null) return repeat while I next == null I next locked := false
Algorithm 2 – Simple RW lock with reader preference Bit 0 – writer active?Bit 31:1 – count of interested readers start_write – repeat until compare_and_swap(L,0, 0x1) start_read – atomic_add(L,2);repeat until ((L & 0x1) = 0) end_write – atomic_add(L, -1) end_read – atomic_add(L, -2)
Algorithm 3 – Fair Lock Writer CountReader Count start_write prev = fetch_clear_then_add(L requests, MASK, 1) // ++ write requests repeat until completions = prev // wait for previous readers and writers to go first end_write – clear_then_add(L completions, MASK,1) // ++ write completions start_read // ++ read request, get count of prev writers prev_writer = fetch_clear_then_add(L requests, MASK, 1) & MASK repeat until (completions & MASK) = prev_writer // wait for prev writers to go first end_read – clear_then_add(L completions, MASK,1) // ++ read completions Requests Writer CountReader Count Completions
So far so good, but … Algorithm 2 and 3 spin on a shared memory location. What we want is for the algorithms to spin on processor local variables. Note – results weren’t presented for Algorithms 2 and 3. We can guess the performance though, since we know the general characteristics of contention.
Algorithm 4 Fair R/W Lock: Local-Only Spinning Fairness Algorithm –read request granted when all previous write requests have completed –write request granted when all previous read and write requests have completed
Case 1: Just a Read Pred == nil Lock.tail I Upon exit: Lock.tail I Lock.reader_count == 1
Case 1: Exit Read next == nil Lock.tail I, so cas ret T Lock.reader_count == 1 Lock.next_writer == nil Upon Exit: Lock.tail == nil Lock.reader_count == 0
Case 2: Overlapping Read After first read: Lock.tail I 1 Lock.reader_count == 1 not nil !!!! pred class == reading Pred->state == [false,none] Locked.reader_count == 2
Case 2: Overlapping Read After the 2nd read enters: Locked.tail I 2 I 1 next == I 2
Case 2: Overlapping reads I1 finishes next != nil I2 finishes Locked.tail = nil count goes to zero after I1 and I2 finish
Case 3: Read Overlaps Write The previous cases weren’t interesting, but they did help us get familiar with the data structures and (some of) the code. Now we need to consider the case where a “write” has started, but a read is requested. The read should block (spin) until the write completes. We need to “prove” that the spinning occurs on a locally cached memory location.
Case 3: Read Overlaps Write The Write Upon exit: Locked.tail I Locked.next_writer = nil I.class = writing, I.next = nil I.blocked = false, success… = none pred == nil reset blocked to false
Case 3: Read Overlaps Write The Read pred class == writing wait here for write to complete
Case 3: Read Overlaps Write The Write Completes I.next The Read Yes! Works, but is “uncomfortable” because concerns aren’t separated unlock the reader
Case 3: What if there were more than 1 reader? change the predecessor reader wait here Yes! Changed by the successor unblock the successor
Case 4: Write Overlaps Read Overlapping reads form a chain The overlapping write, “spins” waiting for the read chain to complete Reads that “enter” after the write as “enter”, but before the write completes (even while the write is “spinning”), form a chain following the write (as with case 3).
Algorithm 5 Reader Preference R/W Local-Only Spinning We’ll look at the Reader-Writer-Reader case and demonstrate that the second Reader completes before the Writer is signaled to start.
1 st Reader ++reader_count Waflag == 0 false 1 st reader just runs!
Overlapping Write queue the write Register writer interest, result not zero, since there is a reader We have a reader, so the cas fails. The writer blocks here waiting for a reader set blocked = false
2 nd Reader Still no active reader ++reader_count
Reader Completes Only last reader will satisfy equality Last reader to complete will set WAFLAG and unblock writer
Algorithm 6 Writer Preference R/W Local-Only Spinning We’ll look at the Writer-Reader-Writer case and demonstrate that the second Writer completes before the Reader is signaled to start.
Last Writer Completes clear write flags signal readers
Unblock Readers ++reader count, clear rdr’s interested no writers waiting or active empty the “waiting” reader list when this reader continues, it will unblock the “next” reader -- which will unblock the “next” reader, etc. reader count gets bumped
Results & Conclusion The authors reported results for a different algorithm than was presented here. The “algorithms” used were “more” costly in a multiprocessor environment; so they’re claiming that the algorithms presented here would be “better.”
Timing Results Latency is costly because of the number of atomic operations.