1  2004 Morgan Kaufmann Publishers Lecture 11, Oct. 29, 2007 Syllabus change: –Tues, Oct. 29, 2007 – Caching –Thurs, Nov. 1, 2007 – Review –Tues, Nov. 6, 2007 – Midterm –Tues, Nov. 15, 2007 – Quiz (caching/V-M) (%10) –Thurs, Nov. 29, 2007 – Final Project / Final Day of Class Still PopNet –Make modifications to the code Add lasers to simulated code Add no virtual memory, but make the I/O BW large and scalable Add network coding/computation at each node (ECC, compression, etc) I will give HW on Caching/Virtual Memory (Due Nov. 15, same day as quiz occurs)

2  2004 Morgan Kaufmann Publishers Chapter Seven

3  2004 Morgan Kaufmann Publishers SRAM: –value is stored on a pair of inverting gates –very fast but takes up more space than DRAM (4 to 6 transistors) DRAM: –value is stored as a charge on capacitor (must be refreshed) –very small but slower than SRAM (factor of 5 to 10) Memories: Review

4  2004 Morgan Kaufmann Publishers Users want large and fast memories! SRAM access times are.5 – 5ns at cost of $4000 to $10,000 per GB. DRAM access times are 50-70ns at cost of $100 to $200 per GB. Disk access times are 5 to 20 million ns at cost of $.50 to $2 per GB. Try and give it to them anyway –build a memory hierarchy Exploiting Memory Hierarchy 2004

5  2004 Morgan Kaufmann Publishers Locality A principle that makes having a memory hierarchy a good idea If an item is referenced, temporal locality: it will tend to be referenced again soon spatial locality: nearby items will tend to be referenced soon. Why does code have locality? Our initial focus: two levels (upper, lower) –block: minimum unit of data –hit: data requested is in the upper level –miss: data requested is not in the upper level

6  2004 Morgan Kaufmann Publishers Two issues: –How do we know if a data item is in the cache? –If it is, how do we find it? Our first example: – block size is one word of data – "direct mapped" For each item of data at the lower level, there is exactly one location in the cache where it might be. e.g., lots of items at the lower level share locations in the upper level Cache

7  2004 Morgan Kaufmann Publishers Mapping: address is modulo the number of blocks in the cache Direct Mapped Cache

8  2004 Morgan Kaufmann Publishers For MIPS: What kind of locality are we taking advantage of? Direct Mapped Cache

9  2004 Morgan Kaufmann Publishers Taking advantage of spatial locality: Direct Mapped Cache

10  2004 Morgan Kaufmann Publishers Read hits –this is what we want! Read misses –stall the CPU, fetch block from memory, deliver to cache, restart Write hits: –can replace data in cache and memory (write-through) –write the data only into the cache (write-back the cache later) Write misses: –read the entire block into the cache, then write the word Hits vs. Misses

11  2004 Morgan Kaufmann Publishers Make reading multiple words easier by using banks of memory It can get a lot more complicated... Hardware Issues

12  2004 Morgan Kaufmann Publishers Increasing the block size tends to decrease miss rate: Use split caches because there is more spatial locality in code: Performance

13  2004 Morgan Kaufmann Publishers Performance Simplified model: execution time = (execution cycles + stall cycles)  cycle time stall cycles = # of instructions  miss ratio  miss penalty Two ways of improving performance: –decreasing the miss ratio –decreasing the miss penalty What happens if we increase block size?

14  2004 Morgan Kaufmann Publishers Compared to direct mapped, give a series of references that: –results in a lower miss ratio using a 2-way set associative cache –results in a higher miss ratio using a 2-way set associative cache assuming we use the “least recently used” replacement strategy Decreasing miss ratio with associativity

15  2004 Morgan Kaufmann Publishers An implementation

16  2004 Morgan Kaufmann Publishers Performance

17  2004 Morgan Kaufmann Publishers Decreasing miss penalty with multilevel caches Add a second level cache: –often primary cache is on the same chip as the processor –use SRAMs to add another cache above primary memory (DRAM) –miss penalty goes down if data is in 2nd level cache Example: –CPI of 1.0 on a 5 Ghz machine with a 5% miss rate, 100ns DRAM access –Adding 2nd level cache with 5ns access time decreases miss rate to.5% Using multilevel caches: –try and optimize the hit time on the 1st level cache –try and optimize the miss rate on the 2nd level cache

18  2004 Morgan Kaufmann Publishers Cache Complexities Not always easy to understand implications of caches: Theoretical behavior of Radix sort vs. Quicksort Observed behavior of Radix sort vs. Quicksort

19  2004 Morgan Kaufmann Publishers Cache Complexities Here is why: Memory system performance is often critical factor –multilevel caches, pipelined processors, make it harder to predict outcomes –Compiler optimizations to increase locality sometimes hurt ILP Difficult to predict best algorithm: need experimental data

20  2004 Morgan Kaufmann Publishers Virtual Memory Main memory can act as a cache for the secondary storage (disk) Advantages: –illusion of having more physical memory –program relocation –protection

21  2004 Morgan Kaufmann Publishers Pages: virtual memory blocks Page faults: the data is not in memory, retrieve it from disk –huge miss penalty, thus pages should be fairly large (e.g., 4KB) –reducing page faults is important (LRU is worth the price) –can handle the faults in software instead of hardware –using write-through is too expensive so we use writeback

22  2004 Morgan Kaufmann Publishers Page Tables

23  2004 Morgan Kaufmann Publishers Page Tables

24  2004 Morgan Kaufmann Publishers (can stop here) Making Address Translation Fast A cache for address translations: translation lookaside buffer Typical values: entries, miss-rate:.01% - 1% miss-penalty: 10 – 100 cycles

25  2004 Morgan Kaufmann Publishers TLBs and caches

26  2004 Morgan Kaufmann Publishers TLBs and Caches

27  2004 Morgan Kaufmann Publishers Modern Systems

28  2004 Morgan Kaufmann Publishers Modern Systems Things are getting complicated!

29  2004 Morgan Kaufmann Publishers Processor speeds continue to increase very fast — much faster than either DRAM or disk access times Design challenge: dealing with this growing disparity –Prefetching? 3 rd level caches and more? Memory design? Some Issues

30  2004 Morgan Kaufmann Publishers Chapters 8 & 9 (partial coverage)

31  2004 Morgan Kaufmann Publishers Interfacing Processors and Peripherals I/O Design affected by many factors (expandability, resilience) Performance: — access latency — throughput — connection between devices and the system — the memory hierarchy — the operating system A variety of different users (e.g., banks, supercomputers, engineers)

32  2004 Morgan Kaufmann Publishers I/O Important but neglected “The difficulties in assessing and designing I/O systems have often relegated I/O to second class status” “courses in every aspect of computing, from programming to computer architecture often ignore I/O or give it scanty coverage” “textbooks leave the subject to near the end, making it easier for students and instructors to skip it!” GUILTY! — we won’t be looking at I/O in much detail — be sure and read Chapter 8 in its entirety. — you should probably take a networking class!

33  2004 Morgan Kaufmann Publishers I/O Devices Very diverse devices — behavior (i.e., input vs. output) — partner (who is at the other end?) — data rate

34  2004 Morgan Kaufmann Publishers I/O Example: Disk Drives To access data: — seek: position head over the proper track (3 to 14 ms. avg.) — rotational latency: wait for desired sector (.5 / RPM) — transfer: grab the data (one or more sectors) 30 to 80 MB/sec

35  2004 Morgan Kaufmann Publishers I/O Example: Buses Shared communication link (one or more wires) Difficult design: — may be bottleneck — length of the bus — number of devices — tradeoffs (buffers for higher bandwidth increases latency) — support for many different devices — cost Types of buses: — processor-memory (short high speed, custom design) — backplane (high speed, often standardized, e.g., PCI) — I/O (lengthy, different devices, e.g., USB, Firewire) Synchronous vs. Asynchronous — use a clock and a synchronous protocol, fast and small but every device must operate at same rate and clock skew requires the bus to be short — don’t use a clock and instead use handshaking

36  2004 Morgan Kaufmann Publishers I/O Bus Standards Today we have two dominant bus standards:

37  2004 Morgan Kaufmann Publishers Other important issues Bus Arbitration: — daisy chain arbitration (not very fair) — centralized arbitration (requires an arbiter), e.g., PCI — collision detection, e.g., Ethernet Operating system: — polling — interrupts — direct memory access (DMA) Performance Analysis techniques: — queuing theory — simulation — analysis, i.e., find the weakest link (see “I/O System Design”) Many new developments

38  2004 Morgan Kaufmann Publishers Pentium 4 I/O Options

39  2004 Morgan Kaufmann Publishers Fallacies and Pitfalls Fallacy: the rated mean time to failure of disks is 1,200,000 hours, so disks practically never fail. Fallacy: magnetic disk storage is on its last legs, will be replaced. Fallacy: A 100 MB/sec bus can transfer 100 MB/sec. Pitfall: Moving functions from the CPU to the I/O processor, expecting to improve performance without analysis.

40  2004 Morgan Kaufmann Publishers Multiprocessors Idea: create powerful computers by connecting many smaller ones good news: works for timesharing (better than supercomputer) bad news: its really hard to write good concurrent programs many commercial failures

41  2004 Morgan Kaufmann Publishers Questions How do parallel processors share data? — single address space (SMP vs. NUMA) — message passing How do parallel processors coordinate? — synchronization (locks, semaphores) — built into send / receive primitives — operating system protocols How are they implemented? — connected by a single bus — connected by a network

42  2004 Morgan Kaufmann Publishers Supercomputers Plot of top 500 supercomputer sites over a decade:

43  2004 Morgan Kaufmann Publishers Using multiple processors an old idea Some SIMD designs: Costs for the the Illiac IV escalated from $8 million in 1966 to $32 million in 1972 despite completion of only ¼ of the machine. It took three more years before it was operational! “For better or worse, computer architects are not easily discouraged” Lots of interesting designs and ideas, lots of failures, few successes

44  2004 Morgan Kaufmann Publishers Topologies

45  2004 Morgan Kaufmann Publishers Clusters Constructed from whole computers Independent, scalable networks Strengths: –Many applications amenable to loosely coupled machines –Exploit local area networks –Cost effective / Easy to expand Weaknesses: –Administration costs not necessarily lower –Connected using I/O bus Highly available due to separation of memories In theory, we should be able to do better

46  2004 Morgan Kaufmann Publishers Google Serve an average of 1000 queries per second Google uses 6,000 processors and 12,000 disks Two sites in silicon valley, two in Virginia Each site connected to internet using OC48 (2488 Mbit/sec) Reliability: –On an average day, 20 machines need rebooted (software error) –2% of the machines replaced each year In some sense, simple ideas well executed. Better (and cheaper) than other approaches involving increased complexity

47  2004 Morgan Kaufmann Publishers Concluding Remarks Evolution vs. Revolution “More often the expense of innovation comes from being too disruptive to computer users” “Acceptance of hardware ideas requires acceptance by software people; therefore hardware people should learn about software. And if software people want good machines, they must learn more about hardware to be able to communicate with and thereby influence hardware engineers.”