TK 2123 COMPUTER ORGANISATION & ARCHITECTURE Lecture 7: CPU and Memory (3)

Contents This lecture will discuss: Cache. Error Correcting Codes.

The Memory Hierarchy Trade-off: cost, capacity and access time. Faster access time, greater cost per bit. Greater capacity, smaller cost per bit. Greater capacity, slower access time. Access time - the time it takes to perform a read or write operation. Memory Cycle time -Time may be required for the memory to “recover” before next access, i.e. access + recovery. Transfer Rate - rate at which data can be moved.

A five-level memory hierarchy. Memory Hierarchies A five-level memory hierarchy.

Hierarchy List Registers L1 Cache L2 Cache Main memory Disk cache Disk Optical Tape decreasing cost/bit, increasing capacity, and slower access time Internal memory external memory

Hierarchy List It would be nice to use only the fastest memory, but because that is the most expensive memory, we trade off access time for cost by using more of the slower memory. The design challenge is to organise the data and programs in memory so that the accessed memory words are usually in the faster memory. In general, it is likely that most future accesses to main memory by the processor will be to locations recently accessed. So the cache automatically retains a copy of some of the recently used words from the DRAM. If the cache is designed properly, then most of the time the processor will request memory words that are already in the cache.

Hierarchy List No one technology is optimal in satisfying the memory requirements for a computer system. As a consequence, the typical computer system is equipped with a hierarchy of memory subsystems; some internal to the system (directly accessible by the processor) and some external (accessible by the processor via an I/O module).

Cache Small amount of fast memory Sits between normal main memory and CPU May be located on CPU chip or module or cache line.

Cache The cache contains a copy of portions of main memory. When the processor attempts to read a word of memory, a check is made to determine if the word is in the cache. If so (hit), the word is delivered to the processor. If not (miss), a block of main memory, consisting of some fixed number of words, is read into the cache and then the word is delivered to the processor. Because of the phenomenon of locality of reference, when a block of data is fetched into the cache to satisfy a single memory reference, it is likely that there will be future references to that same memory location or to other words in the block. The ratio of hits to the total number of requests is known as the hit ratio.

Cache/Main Memory Structure

Cache operation – overview CPU requests contents of memory location Check cache for this data If present, get from cache (fast) If not present, read required block from main memory to cache Then deliver from cache to CPU Cache includes tags to identify which block of main memory is in each cache slot

Cache Operation

Cache Design Size Mapping Function Replacement Algorithm Write Policy Block Size Number of Caches – L1, L2, L3 etc.

Size does matter Cost More cache is expensive Speed More cache is faster (up to a point) Checking cache for data takes time We would like the size of the cache to be small enough so that the overall average cost per bit is close to that of main memory alone and large enough so that the overall average access time is close to that of the cache alone. The larger the cache, the larger the number of gates involved in addressing the cache. The result is that large caches tend to be slightly slower than small ones.

Comparison of Cache Sizes   Comparison of Cache Sizes Processor Type Year of Introduction L1 cachea L2 cache L3 cache IBM 360/85 Mainframe 1968 16 to 32 KB — PDP-11/70 Minicomputer 1975 1 KB VAX 11/780 1978 16 KB IBM 3033 64 KB IBM 3090 1985 128 to 256 KB Intel 80486 PC 1989 8 KB Pentium 1993 8 KB/8 KB 256 to 512 KB PowerPC 601 32 KB PowerPC 620 1996 32 KB/32 KB PowerPC G4 PC/server 1999 256 KB to 1 MB 2 MB IBM S/390 G4 1997 256 KB IBM S/390 G6 8 MB Pentium 4 2000 IBM SP High-end server/ supercomputer 64 KB/32 KB CRAY MTAb Supercomputer Itanium 2001 16 KB/16 KB 96 KB 4 MB SGI Origin 2001 High-end server Itanium 2 2002 6 MB IBM POWER5 2003 1.9 MB 36 MB CRAY XD-1 2004 64 KB/64 KB 1MB   a Two values seperated by a slash refer to instruction and data caches b Both caches are instruction only; no data caches

Cache: Mapping Function Cache lines < main memory blocks: An algorithm is needed for mapping main memory blocks into cache lines. Three techniques: Direct Associative set associative

Direct Mapping Each block of main memory maps to only one cache line i.e. if a block is in cache, it must be in one specific place. pros & cons Simple Inexpensive Fixed location for given block If a program accesses 2 blocks that map to the same line repeatedly, cache misses are very high

Associative Mapping A main memory block can load into any line of cache Memory address is interpreted as tag and word Tag uniquely identifies block of memory Every line’s tag is examined for a match Disadvantage: Cache searching gets expensive The complex circuitry is required to examine the tags of all cache lines in parallel.

Set Associative Mapping A compromise that exhibits the strengths of both the direct and associative approaches while reducing their disadvantages. Cache is divided into a number of sets. Each set contains a number of lines. A given block maps to any line in a given set e.g. Block B can be in any line of set i. With fully associative mapping, the tag in a memory address is quite large and must be compared to the tag of every line in the cache. With k-way set associative mapping, the tag in a memory address is much smaller and is only compared to the k tags within a single set.

Replacement Algorithms When cache memory is full, some block in cache memory must be selected for replacement. Direct mapping : No choice Each block only maps to one line Replace that line

Replacement Algorithms (2) Associative & Set Associative Hardware implemented algorithm (speed) Least Recently used (LRU) An LRU algorithm, keeps track of the usage of each block and replaces the block that was last used the longest time ago. First in first out (FIFO) replace block that has been in cache longest Least frequently used (LFU) replace block which has had fewest hits Random

Write Policy Issues: Must not overwrite a cache block unless main memory is up to date Multiple CPUs may have individual caches I/O may address main memory directly

Write through All writes go to main memory as well as cache Multiple CPUs can monitor main memory traffic to keep local (to CPU) cache up to date Disadvantage: Lots of traffic Slows down writes Create a bottleneck.

Cache: Line Size As the block size increases from very small to larger sizes, the hit ratio will at first increase because of the principle of locality. Two issues: Larger blocks reduce the number of blocks that fit into a cache. Because each block fetch overwrites older cache contents, a small number of blocks results in data being overwritten shortly after they are fetched. As a block becomes larger, each additional word is farther from the requested word, therefore less likely to be needed in the near future.

Number of Caches external cache: Is it still desirable? Multilevel Caches: On-chip cache: A cache on the same chip as the processor. Reduces the processor’s external bus activity and therefore speeds up execution times and increases overall system performance. external cache: Is it still desirable? Yes - most contemporary designs include both on-chip and external caches. E.g. two-level cache, with the internal cache (L1) and the external cache (L2). Why? If there is no L2 cache and the processor makes an access request for a memory location not in the L1 cache, then the processor must access DRAM or ROM memory across the bus – poor performance.

Number of Caches More recently, it has become common to split the cache into two: one dedicated to instructions and one dedicated to data. There are two potential advantages of a unified cache: For a given cache size, a unified cache has a higher hit rate than split caches because it balances the load between instruction and data fetches automatically. Only one cache needs to be designed and implemented. The trend is toward split caches, such as the Pentium and PowerPC, which emphasize parallel instruction execution and the prefetching of predicted future instructions. Advantage: It eliminates contention for the cache between the instruction fetch/decode unit and the execution unit.

Processor on which feature first appears Intel Cache Evolution Problem Solution Processor on which feature first appears External memory slower than the system bus. Add external cache using faster memory technology. 386 Increased processor speed results in external bus becoming a bottleneck for cache access. Move external cache on-chip, operating at the same speed as the processor. 486 Internal cache is rather small, due to limited space on chip Add external L2 cache using faster technology than main memory Contention occurs when both the Instruction Prefetcher and the Execution Unit simultaneously require access to the cache. In that case, the Prefetcher is stalled while the Execution Unit’s data access takes place. Create separate data and instruction caches. Pentium Increased processor speed results in external bus becoming a bottleneck for L2 cache access. Create separate back-side bus that runs at higher speed than the main (front-side) external bus. The BSB is dedicated to the L2 cache. Pentium Pro Move L2 cache on to the processor chip. Pentium II Some applications deal with massive databases and must have rapid access to large amounts of data. The on-chip caches are too small. Add external L3 cache. Pentium III   Move L3 cache on-chip. Pentium 4

Locality Page 129: Stalling.

Internal Memory (revision)

Memory Packaging and Types A group of chips, typically 8 or 16, is mounted on a tiny PCB and sold as a unit. SIMM - single inline memory module, has a row of connectors on one side. DIMM – Dual inline memory module, has a row of connectors on both side. A SIMM holding 256 MB. Two of the chips control the SIMM.

Error Correction Hard Failure Soft Error Permanent defect Caused by harsh environmental abuse, manufacturing defects, and wear. Soft Error Random, non-destructive No permanent damage to memory Caused by power supply problems. Detected using Hamming error correcting code.

Error Correction When reading out the stored word, a new set of K code bits is generated from M data bits and compared with fetch code bits. Results: No errors – the fetch data bits are sent out. An error is detected, and it is possible to correct the error. Data bits + error correction bits corrector sent out the corrected set of M bits. An error is detected, but it is not possible to correct the error. This condition is reported.

Error Correcting Code Function A function to produce code Stored codeword: M+K bits

Error Correction Page 73, 74, 75 Tanenbaum.

Error Correcting Codes (1) Number of check bits for a code that can correct a single error

Error Correcting Codes (2) (a) Encoding of 1100 (b) Even parity added (c) Error in AC

Error Correcting Codes (3) Construction of the Hamming code for the memory word 11110000010101110 by adding 5 check bits to the 16 data bits.

Thank you Q & A