數(shù)據(jù)結(jié)構(gòu)教學(xué)課件：chapter8

1、120:452Primary storage: Main memory (RAM)Secondary Storage: Peripheral devicesDisk drivesTape drives20:453RAM is usually volatile.RAM is about 1/4 million times faster than disk.MediumEarly 1996 Mid 1997 Early 2000RAM$45.007.001.50Disk1Floppy0.500.360.25Tape0.030.010.00120:454Minimize the

2、 number of disk accesses!1. Arrange information so that you get what you want with few disk accesses.2. Arrange information to minimize future disk accesses.An organization for data on disk is often called a file structure.Disk-based space/time tradeoff: Compress information to save processing time

3、by reducing disk accesses.20:45520:456A sector is the basic unit of I/O.Interleaving factor: Physical distance between logically adjacent sectors on a track.20:457Locality of Reference: When record is read from disk, next request is likely to come from near the same place in the file.Cluster: Smalle

4、st unit of file allocation, usually several sectors.Extent: A group of physically contiguous clusters.Internal fragmentation: Wasted space within sector if record size does not match sector size; wasted space within cluster if file size is not a multiple of cluster size.Sector headers: Contains the

5、sector address and an error detection code for the contents of that sector. See P267 Figure 8.5.20:458Seek time: Time for I/O head to reach desired track. Largely determined by distance between I/O head and desired track.Track-to-track time: Minimum time to move from one track to an adjacent track.A

6、verage Seek time: Average time to reach a track for random access.20:459Rotational Delay or Latency: Time for data to rotate under I/O head.One half of a rotation on average.At 7200 rpm, this is 8.3/2 = 4.2ms.Transfer time: Time for data to move under the I/O head.At 7200 rpm: Number of sectors read

7、/Number of sectors per track * 8.3ms.20:451016.8 GB disk on 10 platters = 1.68GB/platter13,085 tracks/platter256 sectors/track512 bytes/sectorTrack-to-track seek time: 2.2 msAverage seek time: 9.5ms8 sector/clusters =4KB, 32 clusters/track.Interleaving factor of 3.5400RPM20:4511Read a 1MB file divid

8、ed into 2048 records of 512 bytes (1 sector) each.Assume all records are on 8 contiguous tracks.First track: 9.5 + 11.1/2 + 3 x 11.1 = 48.4 msRemaining 7 tracks: 2.2 + 11.1/2 + 3 x 11.1 = 41.1 ms.Total: 48.4 + 7 * 41.1 = 335.7ms20:4512Read a 1MB file divided into 2048 records of 512 bytes (1 sector)

9、 each.Assume all file clusters are randomly spread across the disk.256 clusters. Cluster read time is (3 x 8)/256 of a rotation for about 1 ms.256(9.5 + 11.1/2 + (3 x 8)/256) is about 4119.2 ms. 20:4513Read time for one track:9.5 + 11.1/2 + 3 x 11.1 = 48.4ms.Read time for one sector:9.5 + 11.1/2 + (

10、1/256)11.1 = 15.1ms.Read time for one byte:9.5 + 11.1/2 = 15.05 ms.Nearly all disk drives read/write one sector at every I/O access.Also referred to as a page.20:4514The information in a sector is stored in a buffer or cache.If the next I/O access is to the same buffer, then no need to go to disk.Th

11、ere are usually one or more input buffers and one or more output buffers.20:4515A series of buffers used by an application to cache disk data is called a buffer pool.Virtual memory uses a buffer pool to imitate greater RAM memory by actually storing information on disk and “swapping” between disk an

12、d RAM.20:451620:4517Which buffer should be replaced when new data must be read?First-in, First-out: Use the first one on the queue.Least Frequently Used (LFU): Count buffer accesses, reuse the least used.Least Recently used (LRU): Keep buffers on a linked list. When buffer is accessed, bring it to f

13、ront. Reuse the one at end.20:4518class BufferPool / (1) Message Passingpublic: virtual void insert(void* space, int sz, int pos) = 0; virtual void getbytes(void* space, int sz, int pos) = 0;class BufferPool / (2) Buffer Passingpublic: virtual void* getblock(int block) = 0; virtual void dirtyblock(i

14、nt block) = 0; virtual int blocksize() = 0;20:4519Problem: Sorting data sets too large to fit into main memory.Assume data are stored on disk drive.To sort, portions of the data must be brought into main memory, processed, and returned to disk.An external sort should minimize disk accesses.20:4520Se

15、condary memory is divided into equal-sized blocks (512, 1024, etc)A basic I/O operation transfers the contents of one disk block to/from main memory.Under certain circumstances, reading blocks of a file in sequential order is more efficient. Primary goal is to minimize I/O operations.Assume only one

16、 disk drive is available.20:4521Often, records are large, keys are small.Ex: Payroll entries keyed on ID numberApproach 1: Read in entire records, sort them, then write them out again.Approach 2: Read only the key values, store with each key the location on disk of its associated record.After keys a

17、re sorted the records can be read and rewritten in sorted order.20:4522Quicksort requires random access to the entire set of records.Better: Modified Mergesort algorithm.Process n elements in Q(log n) passes.A group of sorted records is called a run.20:4523Split the file into two run files.Read in a

18、 block from each run file.Take first record from each block, output them in sorted order.Take next record from each block, output them to a second file in sorted order.Repeat until finished, alternating between output files. Read new input blocks as needed.Repeat steps 2-5, except this time input fi

19、les have runs of two sorted records that are merged together.Each pass through the files provides larger runs.20:452420:4525How can we reduce the number of Mergesort passes? Not use Mergesort on small runs, read in a block of data and sort it in memory, and then output it as a single sorted run. (Se

20、e example in page 280)In general, external sorting consists of two phases:Break the files into large initial runsMerge the runs together to form a single sorted run file.20:4526General approaches:Read as much of the data of file into memory as possible.Perform an in-memory sort such as Quicksort or

21、Heapsort.Output this group of sorted records as a single run.20:4527A basic I/O operation is read/write a block. Replacement selection sort is a technology of generating a run as larger as possible based on the basic I/O operations.It is a slight variation on Heapsort algorithm.Break available memor

22、y into an array for the heap, an input buffer, and an output buffer.20:4528 1) Fill the array from disk. Set LAST=M-1; 2) Build a min-heap; 3) Repeat until the array is empty: a) Send the record with the minimum key value (the root) to the output buffer. b) Let R be the next record in the input buff

23、er. If Rs key value is greater than the key value just output, thenReplace the root with this keyelseReplace the root with the record in array position LAST, and add the next record in the file to a new heap (actually, stick it at the end of the array). c) Sift down the root to recorder the heap.20:

24、4529See Figure 8.9 in page 28420:4530 1) Only the record whose key value is greater than the last key value output can be added to the heap. 2) The record with smaller key value can not be output as a part of the run, but is stored as the data to be handled for the next run; -Heap becomes smaller an

25、d smaller, and the discarded heap space is used to store the records not be output. 3) If the size of available RAM used by array is M, then the smallest run generated by Replacement selection sort is M and the average run is 2M. (Proven process see “Snowplow analogy”)20:4531Simple mergesort: Place runs into two files.Merge the first two ru