Advanced Computer Architecture - Lecture 40: Input/Output systems. This lecture will cover the following: RAID and I/O system design; redundant array of inexpensive disks; I/O benchmarks; I/O system design; service accomplishment; service interruption; network attached storages and reliability;...
CS 704 Advanced Computer Architecture Lecture 40 Input Output Systems (RAID and I/O System Design) Prof Dr M Ashraf Chughtai Today’s Topics Recap: Redundant Array of Inexpensive Disks I/O Benchmarks I/O System Design Conclusion MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) Recap: I/O device’s performance Last time we compared the performance of disk storage and flash memory We noticed that flash is six times faster than the disk for read and the disk is six time faster than the flash for data write – Then we discussed the trends in I/O inter- connects as: the networks, channels and backplanes – The networks offer message-based narrow- pathway for distributed processors over long distance MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) Recap: I/O Interconnects The backplanes offer memory-mapped wide pathway for centralized processing over short distance The interconnects are implemented via buses The buses are classified in two major categories as the I/O bus and CPU-Memory bus The channels are implemented using I/O buses and backplanes using CPU-Memory buses MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) Recap: I/O buses Then we discussed the bus transition protocols which specify the sequence of events and timing requirements in transferring information as synchronous or asynchronous communication We also discussed bus arbitration protocols ― the protocols to reserve the bus by a device that wishes to communicates when multiple devices need the bus access Here, we noticed that the bus arbitration schemes usually try to balance two factors: MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) Recap: I/O System Bus-priority: the device with highest priority should be serviced first Fairness: every device that want to use the bus is guaranteed to get the bus eventually The three bus arbitration schemes are: Daisy Chain Arbitration Centralized Parallel Arbitration Distributed Arbitration MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) Storage I/O Performance Now having discussed the basic types of storage devices and the ways to interconnect them to the CPU, we are going to look into the ways to evaluate the performance of storage I/O systems We know that if a storage device crashes then prime objective of a storage device should be to remember the original information to make storage device reliable The reliability of a system can be improved by using the following four methods MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) Reliability Improvement Fault Avoidance – prevent fault occurrence by construction Fault Tolerance – providing service complying with the service specification by redundancy Error Removal – minimizing the presence of errors by verification Error Forecasting – to estimate the presence, creation and consequence of errors by evaluation MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) Reliability, availability and dependability The performance of storage I/Os is measured in terms of its reliability, availability and dependability These terminologies have been defined by Laprie; in the paper entitled ‘Dependable Computing and Fault Tolerance: Concepts and Terminology; published in the Digest of papers of 15th Annual Symposium on Fault Tolerant Computing (1985) MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) Dependability Laprie defined dependability as the quality of delivered service such that reliance can justifiably be placed on this service; where the service delivered by a system is its observed actual behavior and the system failure occurs when actual behavior deviates from the specified behavior Note that a user perceives a system alternating between two states of delivered service; these states are: MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) 10 RAID 3: Bit-Interleaved Parity Disk Here, every read or write access goes to all the disk For every read access, the parity is computed across recovery group to protect against hard disk failures Note that for the RAID shown here, there is 33% capacity cost for parity However, the wider arrays reduce capacity costs, but decreases expected availability and increases reconstruction time MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) 36 RAID 4:Block-Interleaved Parity and RAID 5:Distributed Block-Interleaved Parity Both the RAID and RAID levels use the same ratio of data disk to parity disk as RAID 3, but they access data differently The distribution of data in RAID verses RAID is shown here In the Block-Interleaved Parity RAID 4, the parity disk is associated to each data block, identical to RAID So it supports a mixture of small read and small writes and large read and large writes MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) 37 RAID 4: Block-Interleaved Parity AAlogical logicalwrite write becomes four becomes four physical physicalI/Os I/Os Independent Independentwrites writes possible because possible becauseof of interleaved parity interleaved parity Reed-Solomon Reed-Solomon Codes Codes("Q") ("Q")for for protection during protection during reconstruction reconstruction D0 D1 D2 D3 P0 D4 D5 D6 D7 P1 D8 D9 D10 D11 P2 D12 D13 D14 D15 P3 D17 D18 D19 P4 D22 D23 P5 D16 D20 MAC/VU-Advanced Computer Architecture D21 Disk Columns Lecture 40 Input / Output System (3) Increasing Logical Disk Addresses Stripe Stripe Unit 38 RAID 4:Block-Interleaved Parity and RAID 5:Distributed Block-Interleaved Parity However, one drawback of this system is that the parity disk must be uploaded on every write, which is bottleneck for back-toback write This bottleneck is resolved in Block interleaved parity RAID 5, where the parity disk is distributed among the blocks Note from the RAID organization shown here that the parity associated each row of the data block is no longer restricted … MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) 39 RAID 5: Distributed Block-Interleaved Parity AAlogical logicalwrite write becomes four becomes four physical physicalI/Os I/Os Independent Independentwrites writes possible because possible becauseof of interleaved parity interleaved parity Reed-Solomon Reed-Solomon Codes Codes("Q") ("Q")for for protection during protection during reconstruction reconstruction MAC/VU-Advanced Computer Architecture D0 D1 D2 D3 P0 D4 D5 D6 P1 D7 D8 D9 P2 D10 D11 D12 P3 D13 D14 D15 P4 D16 D17 D18 D19 D20 D21 D22 D23 P5 Disk Columns Lecture 40 Input / Output System (3) Increasing Logical Disk Addresses Stripe Stripe Unit 40 RAID 4:Block-Interleaved Parity and RAID 5:Distributed Block-Interleaved Parity … to a single disk Hence, this organization allows multiple writes to occur simultaneously as long as the stripe-units are not located in the same disk For example: 1st write to block must also access its parity block P2 (i.e., two reads from two disks – the 1st and 3rd disks) and MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) 41 RAID and RAID 2nd write to block imply an update in P1 (i.e., two reads from two disks – the nd and 4th disks) Thus, the two write could occur at the same time in parallel Where as we you look into the organization of RAID 4, both the P1 and P2 are on the same disk (5th disk) so it would be bottleneck and could not be written simultaneously MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) 42 RAID and RAID In RAID and RAID 5, the parity is stored as blocks and is associated with a set of data blocks In RAID every access goes to all the disks while the levels and use smaller accesses which allow independent access to occur in parallel In RAID and RAID error detection information in each sector is checked independently for ‘small reads’ to see if the data are correct in one sector MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) 43 RAID and RAID While each ‘small write’ would demand that all other disks be accessed to read the rest of information needed to recalculate the parity Let us compare the recalculation of parity on small write for RAID level 3, and Let us assume that we have blocks of data and block of parity The parity calculation of RAID 3, shown here, is straightforward MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) 44 Small Writes update on RAID Old data New data D0' 1.Read D0 2.Read 3.Read D1 D2 D3 P new data + D0' D1 D2 XOR D3 (4 Write) MAC/VU-Advanced Computer Architecture P' (5 Write) Lecture 40 Input / Output System (3) 45 RAID verses RAID and RAID The parity calculation reads blocks D1, D2, and D3 before adding Block D0’ to calculate the new parity P’ Note that here the new data D0 comes directly from CPU, so disk are not involved in reading it The small writes in case of RAID 4/5 are as shown here Here, the old value of D0 is read (1: Read) and compared with new value D0’ to see Which bit will change MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) 46 RAID 4/RAID 5: Small Writes D0 D0' D1 D2 D3 old data (1 Read) new data P old parity (2 Read) + XOR + XOR (4 Write) (3 Write) D0' MAC/VU-Advanced Computer Architecture D1 D2 D3 Lecture 40 Input / Output System (3) P' 47 RAID verses RAID and RAID Once it has been checked, then the old parity P is read and corresponding bits are changed to form P’ This is accomplished by the logical EX-ORs In this example, the disk reads (D1, D2, D3) and disk writes (D0’ and P’) involving all the disks, are replaces with the disk reads (D0,P) and disk writes (D0’, P’), each involving just disks Hence we can say that one (1) Logical Write in RAID and RAID is equivalent to Physical Reads and Physical Writes MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) 48 System Availability: Orthogonal RAIDs Array Controller String Controller String Controller String Controller String Controller String Controller String Controller Data Recovery Group: unit of data redundancy End to End Data Integrity: internal parity protected data paths MAC/VU-Advanced Computer Architecture Redundant Support Components: fans, power supplies, controller, cables Lecture 40 Input / Output System (3) 49 Thanks and Allah Hafiz MAC/VU-Advanced Computer Architecture Lecture 40 Input / Output System (3) 50 ... inter- connects as: the networks, channels and backplanes – The networks offer message-based narrow- pathway for distributed processors over long distance MAC/VU -Advanced Computer Architecture Lecture. .. MAC/VU -Advanced Computer Architecture Lecture 40 Input / Output System (3) 34 RAID 3: Bit-Interleaved Parity Disk logical record 10010011 11001101 10010011 Striped physical records MAC/VU -Advanced. .. MAC/VU -Advanced Computer Architecture Redundant Support Components: fans, power supplies, controller, cables Lecture 40 Input / Output System (3) 49 Thanks and Allah Hafiz MAC/VU -Advanced Computer