ISM 信息存储与管理第二版 and 大话存储2 读书笔记 (1): Host, Connectivity, Disk Drive

Host (Compute) 

1. Memory Virtualization= It creates virtual memory with an address space larger than the physical memory space  

virtual memory manager(VMM). The VMM manages thevirtual-to-physical memory mapping 

A process known aspagingmoves inactive
physical memory pages onto the swap file and brings them back to the physical memory when required.  

moves the least used pages into the swap file so that enough RAM is available for processes that are more active 

2. Device Driver= software that
permits the operating system to interact with a specific device, such as a printer, a mouse, or a disk drive. 

3. Volume Manager = 

In the early days, disk drives appeared to the OS as a number of continuous disk blocks. 

no easy way to scale in or out of the file system’s size.

Logical Volume Managers(LVM)  = layer between the file system
and the physical disk.  partition large disk or aggregate small disks

The basic LVM components arephysical volumes, volume groups, and logical volumes.  

logical volume can be thought of as adisk partition, whereas the volume group itself can bethought of as adisk.  

logical volume is made up of noncontiguous physical extents and mayspan multiple physical volumes.  

4. File System 

A fileis
a collection of related records or data stored as a unit with a name. 

A file systemis a hierarchical
structure of files.  

directories, are containers for storing pointers to multiple files.  

the metadata in a UNIX environment consists of thesuperblock,
theinodes, and the list ofdata blocks free and in use.  

inode is associated with every file and directory and contains information such as thefile
length, ownership, access privileges, time of last access/modification,number of links, and the address of the data. 

Each file system blockis
a contiguous area on the physical disk. The block size of a file system isfixed at the time of its creation.  

A file can span
multiple file system blocks.

Non journaling file systemscause a potential loss of files because they use separate writes
to update their data and metadata.  

system crashes during the write process, the metadata or data might be lost or corrupted. 

system reboots, the file system attempts to update the metadata structures by examining and repairing them.  

A journaling file systemuses
a separate area called alog orjournal.  

enough information in the log to “replay” the log record and complete the operation.  

quick file system check because it looks only at the active, most recently accessed parts of a large filesystem. 

5. Compute Virtualization 

It enables multiple operating systems to run concurrently on single or clustered physical machines.  

achieved by a virtualization layer that resides between thehardware and virtual machines.
This layer is also called thehypervisor.  

Connectivity 

1. Physical Components of Connectivity 

A host interface deviceor
host adapter
connects a host to otherhosts and
storage devices. Examples of host interface devices are host bus adapter (HBA) and network
interface card (NIC). Host bus adaptoris anapplication-specific
integrated circuit(ASIC) board that performs I/O interface functions between the host and storage,
relieving the CPU from additional I/O processing workload. A host typically contains multiple HBAs. 

2. Interface Protocols 

between the host and storage. 

The popular interface protocols used for host to storage communications areIntegrated Device Electronics/Advanced Technology
Attachment(IDE/ATA),Small
Computer System Interface(SCSI),Fibre
Channel(FC) andInternet
Protocol (IP). 

SATA( master&slave device  6 Gb/s )

SCSI supports up to16 devices on a single bus and provides data transfer rates up to 640 MB/s (for the Ultra-640 version). 

Serial attached SCSI (SAS) is a point-to-point serial protocol  (6 Gb/s )

FC 16 Gb/s 

These protocols are implemented on the disk interface controllers.  therefore, disk drivesare known by the protocol interface they support, for example
SATA disk, FCdisk, and so on. 

Disk Drive Performance 

Zoned bit recordinguses the disk
efficiently. 

On older disk drives, the outer tracks had the same number of sectors as the inner tracks, so data density was low on the outer tracks. This was an inefficient use of the available space, as shown
in Figure 2-9 (a). 

Earlier drives used physical addresses consisting of thecylinder, head,and sector(CHS)
number to refer to specific locations on the disk, 

Logical block addressing(LBA) 

simplifies addressing by using alinear address to access physical blocks of data. The
disk controller translates LBA to a CHSaddress, and the host needs to know only the size of the disk drive in terms of the number of blocks. The logical blocks are mapped to physical sectors on a 1:1 basis. 

Seek Time = 

Full Stroke:The time taken by the R/W head to move across the entire width of the disk, from theinnermost
track to the outermost track.

Average:The average time taken by the R/W head to move fromone random track to another, normally
listed as the time for one-third of a full stroke.

Track-to-Track:The time taken by the R/W head to movebetween adjacent tracks. 

Rotational Latency = the time needed to find the sector in one track, average travelling half
round of the track

Average rotational latency for a 15,000 rpm (or 250 rps)drive = 0.5/250 = 2 milliseconds. 

Disk I/O Controller Utilization 

Average response time (TR) = Service time (TS)
/ (1 – Utilization) 



TSis the time taken by the controller
to serve an I/O.

The graph indicates that the response time changes arenonlinear as the utilization increases. When the average
queue sizes are low, the response time remains low. The response time increases slowly with added load on the queue and increases exponentially
when the utilization exceeds 70 percent.