A blog by Emmanuel Goossaert
This is Part 6 over 6 of “Coding for SSDs”. For other parts and sections, you can refer to the Table to Contents. This is a series of articles that I wrote to share what I learned while documenting myself on SSDs, and on how to make code perform well…
This is Part 5 over 6 of “Coding for SSDs”, covering Sections 7 and 8. For other parts and sections, you can refer to the Table to Contents. This is a series of articles that I wrote to share what I learned while documenting myself on SSDs, and on how to make code perform well on SSDs. If you’re in a rush, you can also go directly to Part 6, which is summarizing the content from all the other parts.
Now that I have covered most of the inner workings of solid-state drives in the previous sections, I can provide data that will help build an understanding of which access patterns should be used and why they are indeed better than others. In this part, I explain how writes should be done, how reads should be done, and why concurrent read and write operations are interfering. I also cover a few optimizations at the level of the filesystem which can improve performance.
Translations: This article was translated to Simplified Chinese by Xiong Duo and to Korean by Matt Lee (이 성욱).
This is Part 4 over 6 of “Coding for SSDs”, covering Sections 5 and 6. For other parts and sections, you can refer to the Table to Contents. This is a series of articles that I wrote to share what I learned while documenting myself on SSDs, and on how to make code perform well on SSDs. If you’re in a rush, you can also go directly to Part 6, which is summarizing the content from all the other parts.
In this part, I cover briefly some of the main SSD functionalities such as TRIM and over-provisioning. I am also presenting the different levels of internal parallelism in an SSD, and the concept of clustered block.
Translations: This article was translated to Simplified Chinese by Xiong Duo and to Korean by Matt Lee (이 성욱).
This is Part 3 over 6 of “Coding for SSDs”, covering Sections 3 and 4. For other parts and sections, you can refer to the Table to Contents. This is a series of articles that I wrote to share what I learned while documenting myself on SSDs, and on how to make code perform well on SSDs. If you’re in a rush, you can also go directly to Part 6, which is summarizing the content from all the other parts.
In this part, I am explaining how writes are handled at the page and block level, and I talk about the fundamental concepts of write amplification and wear leveling. Moreover, I describe what is a Flash Translation Layer (FTL), and I cover its two main purposes, logical block mapping and garbage collection. More particularly, I explain how write operations work in the context of a hybrid log-block mapping.
Translations: This article was translated to Simplified Chinese by Xiong Duo and to Korean by Matt Lee (이 성욱).
This is Part 2 over 6 of “Coding for SSDs”, covering Sections 1 and 2. For other parts and sections, you can refer to the Table to Contents. This is a series of articles that I wrote to share what I learned while documenting myself on SSDs, and on how to make code perform well on SSDs. If you’re in a rush, you can also go directly to Part 6, which is summarizing the content from all the other parts.
In this part, I am explaining the basics of NAND-flash memory, cell types, and basic SSD internal architecture. I am also covering SSD benchmarking and how to interpret those benchmarks.
Translations: This article was translated to Simplified Chinese by Xiong Duo and to Korean by Matt Lee (이 성욱).
Translations: This article was translated to Simplified Chinese by Xiong Duo and to Korean by Matt Lee (이 성욱). Introduction I want to make solid-state drives (SSDs) the optimal storage solution for my key-value store project. For that reason, I had to make sure I fully understood how SSDs work,…
