In addition to Weibo, there is also WeChat
Please pay attention
WeChat public account
Shulou
2025-03-04 Update From: SLTechnology News&Howtos shulou NAV: SLTechnology News&Howtos > Internet Technology >
Share
Shulou(Shulou.com)06/01 Report--
This article mainly introduces "what is the concept of server hard disk". In daily operation, I believe that many people have doubts about what the concept of server hard disk is. The editor consulted all kinds of materials and sorted out simple and easy-to-use operation methods. I hope it will be helpful for you to answer the doubts about "what is the concept of server hard disk?" Next, please follow the editor to study!
Sequence and randomness
How do you define sequential and random access to disks? If the logical block address logical block address (LBA) at the beginning of an IO operation is next to the termination LBA of the last IO operation, it is sequential access, otherwise it is random access. The following picture is more https://en.wikipedia.org/wiki/IOPS.
The logical address is the block address recognized in the program, and the numbering order is incremented. In addition to LBA, there is also a PBA (physical block address), and there is a correspondence between general logic and physical addresses. The physical address is the actual address of the data on disk.
Throughput
Throughput refers to the maximum transmission speed that can be achieved when data is read sequentially from disk, usually in MB/s. Throughput is basically used to measure the transmission efficiency of a large amount of data. This should be counted by a single thread.
Latency
It is the time-consuming of an operation, which generally refers to the time between sending the request and returning the result. From the disk point of view, it includes not only the positioning data, but also the time of data transmission. The word latency itself does not specify the amount of data to be accessed.
IOPS
IOPS is the number of operations performed per second, which is generally measured by random access. Various hard disk manufacturers claim that the IOPS of the hard drive is quite high, but this is not the case when actually using it. Because this thing depends on latency and workload, such as the amount of data accessed by an operation, the number of read and write requests waiting in the queue, and so on. In 1998, Intel announced an open source tool for testing IOPS, called Iometer, saying that you should all stop playing by yourselves. We all use this ratio, that is, we set a benchmark, and now we generally use this test.
HDD
HDD is Hard Disk Drive, the hard disk drive, that is, the ordinary mechanical hard drive, which reads and writes data through the magnetic head. There is a difference between random access and sequential access of HDD. I have previously written an article on related topics, index vs full table scan, and review it again today.
Think of the disk as a disc, on which there is a magnetic head for writing, just like a steel pen. To write something in one position, you need to move the magnetic head to that position first. This operation is called seek (rotation + seek), and the next step is to write down the data sequence (write). After writing the data, the head automatically moves backward. If you continue to write next to each other, you don't need seek.
For example, take a Chinese character as a unit, then write a sentence "Hello to the child" in sequence, only need to seek the magnetic head to the position where "small" is to be written at once, and then write "small", and then directly write the "Hello to the child" to the back one by one, and these words all exist together on the disk. Random writing is to store these words separately and not next to each other. Then a total of 6 times of seek,6 and write are required. Sequential writing is always better than random writing, because random writing takes 5 more seek times.
The seek time of a HDD is 10ms, and the throughput is 100MB/s, that is, the data that can be write 100MB per second. So in 1KB, pure write, the data of 1KB only needs 10us, which is 1, 000 of seek. You can see that seek is very unfriendly.
Then you say yes, I can't write it in order. I can't do it without seek. This depends on the specific scene. For example, in Wechat chat, find a friend's dialog box called seek, type text to send is write, when you chat with a girl at the same time, and scold a gay friend, can you not seek?
Because HDD can overwrite, LBA and PBA correspond one to one, and consecutive logical addresses correspond to consecutive physical addresses, so there is a difference between random access and sequential access.
SSD
SSD is solid state drive, solide-state drive, super fast kind. The following figure shows the performance index of a SSD given by Intel.
Sequential access and random access are inherently incomparable, depending on the amount of data per random access. So he didn't use the same unit.
Why write slowly, depends on the structure of SSD, SSD is a three-tier, the highest level is NAND Flash, each NAND Flash has multiple Block, Block also includes a lot of Page,SSD characteristics is to read and write Page as a unit, at least one Page. Usually a Page is 4k or 8k. Cannot overwrite page. To overwrite, you must empty the Block where that page is located. Or you may need to move the data in the block of this page to another erased block. So it's troublesome to write.
Because of the design of these structures, a contiguous piece of LBA on a SSD does not necessarily correspond to a contiguous PBA, and SSD manages this mapping with a flash translation layer Flash Translation Layer (FTL). Therefore, there is no difference between SSD sequential writing and random writing. It can be regarded as SSD that accesses data randomly.
At this point, the study of "what is the concept of server hard disk" is over. I hope to be able to solve your doubts. The collocation of theory and practice can better help you learn, go and try it! If you want to continue to learn more related knowledge, please continue to follow the website, the editor will continue to work hard to bring you more practical articles!
Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.
Views: 0
*The comments in the above article only represent the author's personal views and do not represent the views and positions of this website. If you have more insights, please feel free to contribute and share.
Continue with the installation of the previous hadoop.First, install zookooper1. Decompress zookoope
"Every 5-10 years, there's a rare product, a really special, very unusual product that's the most un
© 2024 shulou.com SLNews company. All rights reserved.