Awsome video! A technical question for you. I am pretty comfortable with SQL DBs. I need to design a data Historian for IoT projects. Does it make sense to use SQL as Historian, or a Historian/time series DB like TimescaleDB is needed?
I am new to the channel, and man you are amazing, keep up the work. I am a first year CS student and I have learnt so much in such a short period of time from you.
very interesting. on the one hand it makes a lot of sense, to have this at the database level. on the otherhand it feels a bit like lockin-in for PostgreSQL, even though PostgreSQL is my favorite
Would also be good to see how you could automatically tackle converting raw intraday prices into split/dividend adjusted for backtesting. Looks like vectorised functions in numpy work best but there’s a lot of ways. Trying to find the fastest, if you have any clues...
Looking forward to all of that! Great breakdown. Pure Python has me curious! I recently looked into a full Python frontend but got pushed straight into react and all the other heavy hitters.
what playlist do you do all this? For some reason I cant find the beginning of this playlist can you post the link please? Also in the future, post your playlists under your video description please. Appreciate your content! -thanks
This is actually pretty common at tech companies (including where I currently work). In reality, I have found most people take about the same amount of vacation as usual (or even less).
Good stuff Larry, have you used / played around with Clickhouse? It's a columnar SQL DB, very fast queries, and compresses well on disk. I just started using it for storing time-series market alerts (like when a stock breaks above N-day high). But it's window functions are lacking / under-development.
I've been collecting all the tick data for the 53 ish pairs available on Kraken. I just store them as csv files on a google drive and do everything in pandas. Is that not the move? Should I be doing this instead though?? Help! im a noob
What about druid? I'm working with a company right now that handles a lot of txns at scale and they prefer Druid. Wondering what do you think about Druid vs TimescaleDB?
Haha, hey Elysse! Didn't expect you to show up in here. That's what happens when I am in quarantine for nearly a year. Now I can see your old art projects on your channel. The one with the penguins + MGMT is adorable :).
Sure, here are the direct links: blog.timescale.com/blog/how-to-store-time-series-data-mongodb-vs-timescaledb-postgresql-a73939734016/ blog.timescale.com/blog/timescaledb-vs-influxdb-for-time-series-data-timescale-influx-sql-nosql-36489299877/ stackshare.io/postgresql
I manage a business that uses time series databases, we just never believed in timescale db it has no “as-of” joins which is very crucial in our trade, quest db basically offers everything timescale db doesn’t as it is also sql based, and compared to something like influx db and kdb + there is just less support for existing API as timescale has a lot of deprecation. That’s just some information, not hating on timescale db but this is why we stay away from it, i was just hoping this video convinced me but still rip to timescale…….
I'm starting a new channel on AI at youtube.com/@parttimeai, please subscribe!
Awsome video! A technical question for you. I am pretty comfortable with SQL DBs. I need to design a data Historian for IoT projects. Does it make sense to use SQL as Historian, or a Historian/time series DB like TimescaleDB is needed?
Ironically, you do a much better job at explaining this technology than the company itself. 👍🏻
You've been missed!! Hope everything is better.
I have to clear up my weekends to sit down and watch your channel, absolutely fascinating stuff that you talk about mate
This is some of the best content I've ever seen on UA-cam.. Thank you Larry!
thanks wolfie, just the information I've been looking for
It is great how you explain the trade-offs for the technological choices you make. Super helpful!
👌🏻welcome back. Tip: use copy_to_table for insertion (much faster than execute or execute_many)
Thanks so much for all of video in your channel. I learned a lot from you.
HELL YES !!! Really excited for this series.
Just found your channel, subscribed, liked and pretty excited, to follow this series. Keep up the great work!
Hi from London ✌
I am new to the channel, and man you are amazing, keep up the work. I am a first year CS student and I have learnt so much in such a short period of time from you.
what a sweet man you are ! thanks for your knowledge sharing.
Just discovered your channel. I'm a beginner with SQL and Python and your channel is awesome. I subscribe !
Welcome! We're just getting started over here, lots of new content planned.
Woot! Let’s get into this!!!
Awesome video! Love from the Timescale team!
Character In the video It's great, I like it a lot $$
YES LARRY feed us your KNAWLEDGES!
very interesting. on the one hand it makes a lot of sense, to have this at the database level. on the otherhand it feels a bit like lockin-in for PostgreSQL, even though PostgreSQL is my favorite
Immutable data? What about "adjusted close" for stock prices? Do you hold an extra table for splits and dividends?
man this is awesome! thanks for the lessons, super excited for this one!
Would also be good to see how you could automatically tackle converting raw intraday prices into split/dividend adjusted for backtesting. Looks like vectorised functions in numpy work best but there’s a lot of ways. Trying to find the fastest, if you have any clues...
Thanks for video mate! does timescale provides clustering with open source version?
Great channel! Do you know any ODBC driver to work with Timescaledb? Thanks!
Looking forward to all of that! Great breakdown. Pure Python has me curious! I recently looked into a full Python frontend but got pushed straight into react and all the other heavy hitters.
what playlist do you do all this? For some reason I cant find the beginning of this playlist can you post the link please? Also in the future, post your playlists under your video description please. Appreciate your content! -thanks
You're probably the most awaited man after Biden to take on to the microphone :)
P.S : I like your podcast mic, what brand is that?
It's a Blue Yeti, one of the most popular mics for podcasts and UA-cam channels since it uses USB and isn't too expensive.
dude, this is fuckin awesome. i'm gonna start following along and build this out. Super cool man
unlimited vacation working at Timescale is the craziest part of this vid
This is actually pretty common at tech companies (including where I currently work). In reality, I have found most people take about the same amount of vacation as usual (or even less).
@@parttimelarry interesting. do deadlines stop people from taking vacation? what is the counter balance?
Love your videos. Do you have one where you query json data saved in Postgresql?
Yay Postgres!!! Done with SQLlite
Hey, can you make one video for predicting 'short squeeze' or an app for screening stocks that have the potential for 'short squeeze'. Thanks.
How does it compare with ClickHouse. It's relational but performance wise seems to beat Timescale.
Can the primary key be a combinatory key of datetime plus asset ID?
Good stuff Larry, have you used / played around with Clickhouse? It's a columnar SQL DB, very fast queries, and compresses well on disk. I just started using it for storing time-series market alerts (like when a stock breaks above N-day high). But it's window functions are lacking / under-development.
Thank you for continuing posting new materials Larry!! 😀 It is very helpful!
Btw, when are you going to update your wordpress website?
Someday :(. I've been up and down for the last couple months so have had trouble writing everything out all the way.
Can't afford to buy you a drink but I will give you a hot stock tip. $GME.
Thanks! Will YOLO some $GME 100C weeklies.
I've been collecting all the tick data for the 53 ish pairs available on Kraken. I just store them as csv files on a google drive and do everything in pandas. Is that not the move? Should I be doing this instead though?? Help! im a noob
Dope
What about druid? I'm working with a company right now that handles a lot of txns at scale and they prefer Druid. Wondering what do you think about Druid vs TimescaleDB?
You are so right about sql vs nosql databases and Postgres specifically. So much anti-relational dogma out there that just isn’t true.
Yay! Crypto, 24/7, if it can only make $4 per hour, that's $96/day, $2880/month
Any particular requests on crypto? Particular cryptocurrency, exchange, strategy, timeframe, etc?
Whaaaaa? Larry with a beard?! What's happening with the world?! 😉🤣
Haha, hey Elysse! Didn't expect you to show up in here. That's what happens when I am in quarantine for nearly a year. Now I can see your old art projects on your channel. The one with the penguins + MGMT is adorable :).
Awwww, thank you! 💖 It's a great song!
I should probably take those art projects down from College. 🤣
@@Purplekaleidoscope77 No way they are cool. What you should do is post more art + music mashups, I think it would be a hit channel.
Thank you Larry! 💖
That does sound pretty dope! I think I've got an idea for another project. Gotta pick up Houdini ( 3d program) again. ☺️
Timescaledb is not working in windows plz guid how to enable it with postgres 15
link blog?
can you post those links in your ppt?
Sure, here are the direct links:
blog.timescale.com/blog/how-to-store-time-series-data-mongodb-vs-timescaledb-postgresql-a73939734016/
blog.timescale.com/blog/timescaledb-vs-influxdb-for-time-series-data-timescale-influx-sql-nosql-36489299877/
stackshare.io/postgresql
I manage a business that uses time series databases, we just never believed in timescale db it has no “as-of” joins which is very crucial in our trade, quest db basically offers everything timescale db doesn’t as it is also sql based, and compared to something like influx db and kdb + there is just less support for existing API as timescale has a lot of deprecation.
That’s just some information, not hating on timescale db but this is why we stay away from it, i was just hoping this video convinced me but still rip to timescale…….
Well what does that business use then, for time-series?
I'd gladly pay you for a bit of help. I know I know.
curious what you do in life ? like who are you ? :)
I supposed I should make some kind of introduction video for the homepage now that there are a good number of people subscribing...
thanks wolfie, just the information I've been looking for