@@ilonachan That'd be more than a lecture. That would be a 4-month course, at the least! And that is assuming you have a solid handle on CS fundamentals and data structures.
like other commenter stated, that would be a multi-month course to get started. To really understand it each of the indexes and how to implement it efficiently, you'd have to spend 1-2 years. If you're short on time, one way to learn how to implement these is to read the source code for postgres, mongodb, cassandra, derby, or sqlite. You can look at the history of those files to see how the implementation evolved.
Really appreciate the effort that goes into making these videos so concise and informative. Glad I found the channel & looking forward to your next episode!
Gotta be honest your content is rock solid already ordered a copy of your book, can we get a video on the solid design pattern? Kinda like the way your animations explain these highly difficult to learn pattern. Thanks!
It will be great if you can come up with a video which maps use cases to DBs. Like which type of DB is best suited for which use cases? Covering popular DBs like RDBMS, Document DB, Key value, Column family DBs, colum oriented DBs, Map reduce, and so on.
!. Skiplist (probabilistic data structure to implement a Sorted Set for efficient lookup, insertion and deletion) ~ B-Tree 2. Hash index: key - hash function -> bucket 3+4. SSTable + LSM Tree (NoSQL) 5. B-Tree (Balanced Tree): Internal nodes to store keys and leave nodes to store actual data (DBMS) 6. Inverted Index: word -> document (Elastic Search) 7. Suffix Tree (Trie): search box 8. R-tree
I LOVE your videos. This is blowing my mind. I'm even an app developer and security specialist and I've never heard of these data structures. This is why I subscribe to ByteByteGo.
Great videos that helps me to understand and enables to further understand what my developers explaining to me. I am from financial services. Curious to know what animation tools are being used, will be useful for our financial services presentations.
Those typically use modified forms of B-tree, R-tree, and structures like red-black trees commonly used as the memtable for LSMs. The in-memory approach either focuses on extreme OLTP performance like the SILO research database and resulting LineairDB embedded database by NTT Data, or in the case of SAP HANA, uses efficiently compressed in-memory columnar B-tree storage for blisteringly fast columnar reads for extremely fast-processing aggregations.
Unless tied to the SAP ecosystem, I'd say stay away from HANA and use a more fitting system: A great open source fault-tolerant OLTP system like FoundationDB, YugabyteDB, or DGraph with modification streaming to a columnar fast-response OLAP system like Apache Druid, previously memory-only but now works on disk too.
Which category does the Vertica Analytic database falls into? And what do you think about the architecture of Vertica database? The concept of logical and physical schema in Vertica is truly unique and beneficial for performance and availability.
Vertica, as far as I'm aware, uses a modified per-column B-tree setup along with bitmap indexes. The column orientation allows for efficient compression and resultant fast bulk reading for analytics workloads, but makes OLTP-type modification and access impractical. An attempt at an HTAP database on these same principles is SAP HANA, which increases the aggressiveness of compression with smaller "chunked" columns residing in memory with a Postgresql-like pending write hold which gets flushed during dips in workload or when many writes are pending on one chunk.
I assume that these data structures are important. The presentation was almost useless, though; far too little was said about how to use these structures, what advantages they have, how they are implemented and so on.
Redis can hold a data structure (for example mentioned sorted under a key) and allows to operate on these data structures without downloading them. That is when other data structures come it, they implement the data structures(lists, sets etc).
@@krajekdev The values associated with a key are user defined, so it's up to the user to choose what data structure it wants in the key's value. Why should Redis dictate that?
@@DK-ox7ze Because we are talking about Redis data structures. As an example, take a look at the SADD command. It operates on a set (internal Redis data structure, not user-defined structure). Other commands operate on lists, hashes etc.
i doubt anything of this powers modern databeses database table is just unsorted array of c structs if you create an array of pointers to rows of table, you can qsort that array of pointers by writing appropriate comparators now you have indexes bsearch them thats all you need to have fun
That might've been the case some time ago, but the video seems to include cases like document storage and analysis, and high-volume tables (millions of rows)... Your approach doesn't seem to be able to handle those needs. 🤓
Subscribe to our weekly newsletter to learn something new every week: bit.ly/3JFtDUK
What software is used to create these presentations :)
after a minute, I do not understand anymore but still watched because of simple cool animation and concise short explanation on each one.
yes, each graphic is probably, at least a month of study ...for me at for me anyway, lol!
Would like to see more on how these are implemented!
I recommend the book "Database internals" for this
This professor is amazing for explaining a lot of these: ruclips.net/video/aZjYr87r1b8/видео.html&ab_channel=AbdulBari
that'd probably be a whole lecture, and I'm here for it
@@ilonachan That'd be more than a lecture. That would be a 4-month course, at the least! And that is assuming you have a solid handle on CS fundamentals and data structures.
like other commenter stated, that would be a multi-month course to get started. To really understand it each of the indexes and how to implement it efficiently, you'd have to spend 1-2 years. If you're short on time, one way to learn how to implement these is to read the source code for postgres, mongodb, cassandra, derby, or sqlite. You can look at the history of those files to see how the implementation evolved.
Amazing job, concise, well presented, right to the point...
Amazing source of tech insights. Thank you for your effort.
Really appreciate the effort that goes into making these videos so concise and informative. Glad I found the channel & looking forward to your next episode!
Amazing visual representations, thank you!
These animations are really awesome. I'm very curious about how they were created
Thank you, sir
Your presentation is great.
Amazing stuff as always
I live those graphics. Not boring to look at, when listening is boring.
Comment section as expected, all appreciated the wonderful work
Gotta be honest your content is rock solid already ordered a copy of your book, can we get a video on the solid design pattern? Kinda like the way your animations explain these highly difficult to learn pattern. Thanks!
It will be great if you can come up with a video which maps use cases to DBs. Like which type of DB is best suited for which use cases? Covering popular DBs like RDBMS, Document DB, Key value, Column family DBs, colum oriented DBs, Map reduce, and so on.
He already did this kind of video ;)
ruclips.net/video/kkeFE6iRfMM/видео.html
He already did: "How To Choose The Right Database?".
@@faroukfaiz8396 Yes but it doesn't cover which database to choose for which scenario. He has only outlined the pros and cons of migrating to a new db
How do you animate your videos?
Your animations are super cool!
Very well presented! Thanks
Please add detailed videos for these algorithms
We would like see detail level video on each item….. with wonderful presentation
Like the animations you create. Can you share what tools you use ? Probably make a video on that :)
Would like to see more details on these data structures!
These video's are amazing, anyone got a clue which program is used to make them?
I wish your videos had 38 million views.
!. Skiplist (probabilistic data structure to implement a Sorted Set for efficient lookup, insertion and deletion) ~ B-Tree
2. Hash index: key - hash function -> bucket
3+4. SSTable + LSM Tree (NoSQL)
5. B-Tree (Balanced Tree): Internal nodes to store keys and leave nodes to store actual data (DBMS)
6. Inverted Index: word -> document (Elastic Search)
7. Suffix Tree (Trie): search box
8. R-tree
I LOVE your videos. This is blowing my mind. I'm even an app developer and security specialist and I've never heard of these data structures. This is why I subscribe to ByteByteGo.
Love the visuals
Nice one, but would really like to see some additional videos in this topic - potentially with more exotic algo and data structures.
What kind of software was used to make the animation? so fascinating
Skiplist, hash index, SSTable, LSM tree, B-tree, Inverted index, Suffix tree, R-tree
Great content
What tools do you use to create this content and the animations
Thanks❤
I am interested in this as well. The Visual aspects of your presentations are just as insightful as your concise verbiage!
Thank you. Can you tell what tool you are using to do this presentation 😊
Can you please cover r-trees and spatial databases? That would be awesome!
Great videos that helps me to understand and enables to further understand what my developers explaining to me. I am from financial services. Curious to know what animation tools are being used, will be useful for our financial services presentations.
I'm curious as well. Great clean visuals.
Awesome video thanks
Would like to see info on hierarchical databases and the various implementations. HIDAM, HDAM, Logical Linking. How HDAM is a 1 I/O access.
would like to see videos on big data, like spark, hdfs, hive etc..
Please do a video on inverted index
Can you also make videos on wavelet trees?
very important concepts
Nice video! what about in memory databases like sap hana? thks for sharing!
Those typically use modified forms of B-tree, R-tree, and structures like red-black trees commonly used as the memtable for LSMs. The in-memory approach either focuses on extreme OLTP performance like the SILO research database and resulting LineairDB embedded database by NTT Data, or in the case of SAP HANA, uses efficiently compressed in-memory columnar B-tree storage for blisteringly fast columnar reads for extremely fast-processing aggregations.
Unless tied to the SAP ecosystem, I'd say stay away from HANA and use a more fitting system: A great open source fault-tolerant OLTP system like FoundationDB, YugabyteDB, or DGraph with modification streaming to a columnar fast-response OLAP system like Apache Druid, previously memory-only but now works on disk too.
Thank you! ❤
Looking for more videos from you on Data structures and Algorithm
Can you please explain all 8 Types in detail video
Can u please create video how you do animations helpful for many you tubers
I want to create my own DBMS. How can I go about it?
What ever it is we are happy.
I would be interested in seeing what data structures graph databases such as Neo4J, etc. use.
While this is a starting point to research those data structures, it's hardly more than that
Which one is used in MS SQL? B-Tree?
Thank you,👏
Which category does the Vertica Analytic database falls into? And what do you think about the architecture of Vertica database? The concept of logical and physical schema in Vertica is truly unique and beneficial for performance and availability.
Vertica, as far as I'm aware, uses a modified per-column B-tree setup along with bitmap indexes. The column orientation allows for efficient compression and resultant fast bulk reading for analytics workloads, but makes OLTP-type modification and access impractical. An attempt at an HTAP database on these same principles is SAP HANA, which increases the aggressiveness of compression with smaller "chunked" columns residing in memory with a Postgresql-like pending write hold which gets flushed during dips in workload or when many writes are pending on one chunk.
👍 1,000th "Like!" Thanks! 😎✌️
Well, you merely _mentioned_ those structures. You cannot call this "discussing" them.
Al Quinn keeps his data in the best of all databases: MS Paint 🖌
Does anyone know what software he uses to build these videos?
I assume that these data structures are important. The presentation was almost useless, though; far too little was said about how to use these structures, what advantages they have, how they are implemented and so on.
No Hash Mapped Array Tries?
And the implementation hell opens its gate once we can't use this structures in memory but have disk based representations.
Awesome
Redis is a key value DB, so Hash index makes perfect sense. But I am wondering why Skip list is used in it?
@@sansmoraxz Why is sorting required when we can access key values in O(1) time?
Redis can hold a data structure (for example mentioned sorted under a key) and allows to operate on these data structures without downloading them. That is when other data structures come it, they implement the data structures(lists, sets etc).
@@krajekdev The values associated with a key are user defined, so it's up to the user to choose what data structure it wants in the key's value. Why should Redis dictate that?
@@DK-ox7ze Because we are talking about Redis data structures. As an example, take a look at the SADD command. It operates on a set (internal Redis data structure, not user-defined structure). Other commands operate on lists, hashes etc.
10/10
I need more details. Here it is too short and shallow. Could be just a matrix instead without sacrificing content.
Make a video of rtrees
i doubt anything of this powers modern databeses
database table is just unsorted array of c structs
if you create an array of pointers to rows of table, you can qsort that array of pointers by writing appropriate comparators
now you have indexes
bsearch them
thats all you need to have fun
That might've been the case some time ago, but the video seems to include cases like document storage and analysis, and high-volume tables (millions of rows)... Your approach doesn't seem to be able to handle those needs. 🤓
You don't have to flex like that.....
cool
Without use cases and time complexity comparsion the Video is useless:/
useless
🤍