I'm absolutely stoked to announce that I am joining
@AmplifyPartners
as an Entrepreneur in Residence and starting my own company with them! In stealth for now, make sure to follow if you want to be the first in line for what I'm cooking up.
After a decade working at
@SingleStoreDB
, it's time for me to move on to the next chapter in my career. Respect and love to everyone at s2, I know you'll continue crushing it without me!
Will be taking a short break before ramping back to full execution speed. Stay tuned. :)
Hey
#Laravel
community! Check out this blog post on getting started with
@SingleStoreDB
and
@laravelphp
written by
@aarondfrancis
!! As always, if you have an awesome project which needs maximum speed from the database - don't hestitate to reach out!
Curious why I've been pretty quiet these last few months? Attend my talk this Thursday (11:45 PT) at
#WasmCon
in person or remote to find out! Live stream available here:
Easily one of the most interesting systems talks this year. TigerBeetle has pulled out all the stops towards building the ultimate dedicated OLTP system. Very impressive! 👏 🎉
“Redesigning OLTP for a New Order of Magnitude” by
@jorandirkgreef
at
@QConSF
is out!
A dense deep dive into TigerBeetle’s:
- network
- storage
- consensus
Plus Online General Purpose Processing (OLGP) vs OLTP, and looking ahead to speculative state machine execution.
Enjoy!
Hope you can join me at WasmCon 2023! I'll be demoing a realtime synchronization engine between Wasm running in the browser and the edge. You won't want to miss this!
#sqlite
#data
#WasmCon
@sched
I just got access (sneak peak) to
@JackEllis
's new course and HOLY CRAP is it good. This is going to be useful for anyone using
@SingleStoreDB
, not just Laravel developers. Also - in classic Jack fashion the course is full of funny easter eggs! 👀👀
With all the recent buzz about
#singlestore
in the
#laravel
community, just want to remind folks we are still looking for people to take our new Laravel driver (by
@aarondfrancis
) for a spin! DM me if you need help getting started!
Last week I sat down with database guru
@carlsverre
to talk about his new project SQLSync.
We cover offline-first, multiplayer, conflict resolution, virtual filesystems, Durable Objects, and (somehow) more!
We got super technical. You're gonna love it.
Interested in offline/local-first? WebAssembly? Multiplayer databases? If so, join in on a conversation I had with
@KateHolterhoff
and
@redmonk
on what a frontend optimized database stack could look like - and why we need it. Thanks for hosting me Kate!
#offlinefirst
#localfirst
@JackEllis
Always happy to nerd out!
@tobias_petry
provided a great start - but lets dive in: First off, because we are a columnstore database adding a new column (with or without a default value) happens almost instantly. This is because each column is stored in separate files.
1/x
Excited to share a reference architecture for scaling global package logistics using
@SingleStoreDB
and Redpanda (
@VectorizedIO
)! Scaled it up to 125x global delivery volume using Google Cloud.
@swyx
@JackEllis
Hey
@swyx
- I agree. We don't advertise our tradeoffs. But of course they exist. Jack's thread is a fantastic read, but if you want to see more please take a look at the following resources.
The first is this Reddit thread in which I try to unpack things:
Shitty Monster Idea:
A goblin with 10d10+30 health. Whenever they take any damage, they are split into another goblin who can take its own turns, but uses the same pile of hp.
#DnD
Is anyone else absolutely pumped for
@JackEllis
's SingleStore for Laravel course?!?! It's going to be absolutely awesome. It's still in presale - grab it now for a discount off full price!
@SingleStoreDB
@laravelphp
Hey gang - announcing a fun project that
@JackEllis
and I are going to put on now and then. We love chatting about technology, and have had some successful webinars in the past. So we figured, why stop now? Our first topic will be on servers vs serverless. Hope to see you there!
Announcing: Jack & Carl
Over the past two years,
@carlsverre
and I have had discussions about various database and scaling challenges. We are now creating a community, powered by a monthly webinar, for CTOs & software engineers at the top of their field.
Github Copilot is so freaking good at automating annoying things like creating a table definition for a csv file. I just type create table, copy in the raw csv header, add a little comment, and BAM.
I just found out that my talk about Revolutionizing Application Architecture with Wasm Powered Database Extensibility was accepted for
#CloudNativeWasmDay
at
#KubeCon
+
#CloudNativeCon
North America 2022! 🎉 Come see me on Oct 24th in Detroit!
This is HUGE for the
@SingleStoreDB
developer community! We will be keeping a close eye and running some of our own testing soon. Nice job
@heychazza
!!
Stoked to see this go out! Also want to send thanks to
@heychazza
for inspiring us to build something like this! Can't wait to see all the awesome Laravel projects built on top of
@SingleStoreDB
Get it while it's hot! SingleStore now has a Docker image which works on Apple Silicon! This will help out with local dev a ton for folks who use a Mac.
Breaking news: SingleStore now runs on M1 Macs. The timing of this release is phenomenal. And this means no more local MySQL for our development environment. I can now play with more than 10,000 database rows locally again 🥳
I'm so excited to see what comes next for
@davidrfgomes
! He was a key member of the SingleStore team and I feel so lucky that I was able to work closely with him over the years. Cheers to an amazing run! 🥂🥳
Today was my last day at
@SingleStoreDB
. I had an amazing final few weeks, and I'm super proud of what we've achieved. Thank you everyone for the camaraderie and for helping me so much along the way!
@JackEllis
@tobias_petry
Of course, adding a column requires allocation, dropping just requires decrementing some ref counts. So to see some seriously crazy numbers - let's drop a column on 21 billion rows:
Tuning up the SingleStore Logistics Simulator today in preparation for a live demo next week - and holy bananas Redpanda's automatic partition rebalancing is slick. Nice job
@VectorizedIO
Is your infra running in the cloud? I bet some of these tweaks apply to your workload. Maybe take a moment to check on cost-saving ideas and send Jack some swag in thanks :)
I’m building SQLSync because I want to make client-side applications easier to build without us having to reinvent the wheel each time. If you share my vision of the future, consider DMing me on Twitter or starring SQLSync on Github.
Thanks for the SQLSync mention and great blog post
@kylemathews
! I agree strongly that it feels like the early days of react. Offline first is a fundamental shift in app architectures that can lead to more robust applications!
New blog post on what I'm observing in the local-first world and where I think it's going. In short, I think it's a big upgrade for the web. It's giving me the same vibes as the React community circa 2014.
If anyone’s looking for a weekend project, the SQLSync guide is up!
It exposes a lot of rough edges which is fantastic cause it means there is tons of room for feedback! Give it a shot and let me know what you would do to make the dev UX 10/10.
It's awesome to see folks like
@paulgb
talk about what I've been calling the rebase model. This is a very similar approach to what SQLSync does in order to keep SQLite databases in sync. One of my favorite parts about this model is that it allows you to model many different kinds
Just finished “The First Fifteen Lives of Harry August” by
@ClaireNorth42
and I can’t recommend it more. Easily the best
#book
I have read in the last year. If you are into thinking about time, you owe it to yourself to check it out!
I'm excited to share that I will be a speaker at
#QConNY
this June 13-15, talking about local-first application architecture. Honored that
@bobcatwilson
included me in her track!
Learn more here:
Ok folks, let’s do a swag roundup. What are you stoked to show off on the daily? For me,
@equinixmetal
nailed it with this bag and amazing recipe books. Joins the ranks with my Rust hoodie and
@SingleStoreDB
beanie!
#CloudNativeCon
#KubeCon
Push compute to data - don’t bring data to compute!
Does this idea inspire you? Are you as excited as I am about the future of data?
Working on some fun stuff at
@SingleStoreDB
. Check out some of our ideas here:
Love
@usefathom
's new high-volume pricing page! Especially the quote "We use the best database software in the business". Proud to count Fathom as one of
@SingleStoreDB
's customers!
Huge congrats due to
@richardartoul
,
@ryanworl
and team! Also check out their latest blog, it’s an excellent analysis of big data infra business models in the cloud.
We raised $20M to build the future of data streaming infrastructure! Thanks to our partners at Amplify and Greylock for believing in us. Read more about our plans in our latest blog post!
Just finished Touch by
@ClaireNorth42
and wow - her writing does not disappoint. Check it out if you like a very unique thriller. Now to figure out what to read next!
New blog post! I wrote about how I use GitHub contributions as another source of motivation when working on solo projects. Specifically I added a mini GitHub contribution graph to my Polybar (system taskbar).
LIVE in 10 minutes! Talking about
@SingleStoreDB
and Redpanda (by
@VectorizedIO
) with
@emaxerrno
!! Hope you will join and ask questions! Will be showing off a live demo of scaling worldwide package logistics to 45x the global volume :)
SingleStore just launched our first ever global database hackathon! If you want to get in on the action, you have the month of October to submit your project for a chance to win part of the $30,000 prize pool!! Register here:
Just landed my first PR on
@excalidraw
! ()
Stoked to contribute to the absolute best visual whiteboarding program on the planet.
Thanks
@dluzar
for the accept :)
Thanks Radu! Super excited to see SQLSync trending! I've been reading that page for years, nice to be a part of it. If anyone is just learning about SQLSync now and have any questions or feedback please don't hesitate to ask!
I'm excited to share that I will be a speaker at
#QConNY
this June 13-15, talking about local-first application architecture. Honored that
@bobcatwilson
included me in her track!
Learn more here:
@JackEllis
@tobias_petry
Secondly, SingleStore speeds up columnstore ingest using a memory accelerated buffer. This buffer is row-oriented (like most OLTP databases) so we need to actually alter each of the rows in the structure. Naively this can be slow, which is why we built online alter table!
2/x
@JackEllis
@tobias_petry
So... why 32 seconds rather than instant? To safely run an online alter in a lock-free distributed system requires coordination with running transactions. Once it can slot itself in safely, the majority of the work happens async.
Let's see how fast it is with no other txns:
@JackEllis
@tobias_petry
Online alter table allows us to commit the schema change before we actually update the datastructure. Then as
@tobias_petry
said, we rebuild rows on read as well as asyncronously in the background to ensure a great user experience!
3/x
People are noticing that ChatGPT can pass various kinds of tests like the Bar exam (lawyer) or tech interviews. Perhaps a signal that we are testing for the wrong thing. Test for problem solving, creativity, and learning. Memorization can be augmented.
@JackEllis
@bradleybernard
Here is my latest thread on the topic which I have now pinned to my profile 😝 as always, if something doesn’t make sense, just ask!
@swyx
@JackEllis
Hey
@swyx
- I agree. We don't advertise our tradeoffs. But of course they exist. Jack's thread is a fantastic read, but if you want to see more please take a look at the following resources.
The first is this Reddit thread in which I try to unpack things:
Playing around with implementing a key/value API for
@SingleStoreDB
. Building a Redis compatible interface was pretty fun. If you want to run KV workloads on SingleStore check it out! Imagine joining your KV data with other data...
👀👀👀
Coming soon: an official
@SingleStoreDB
driver for
@laravelphp
!
Insane SingleStore power and the developer experience you know and love from Laravel.
I still have a few weeks to go, but I can already tell y'all are going to love this.
Can we all just take a moment to appreciate how awesome is? It’s been around for ages yet feels current. Never need to spend more than a few mins to get exactly the information I need. Something we could all strive for!
Working on query subscriptions this week and improving the connection between SQLSync and React. Trying to write down notes and thoughts in issues as I go. Check out the first one here:
#buildinginpublic
#SQLSync
#localfirst
#offlinefirst
Random note: if you get cold working from your office even with a heater on - consider getting an humidifier! Increasing humidity to 50 or 60% makes a huge difference in heat transmission through the room. Major change to my quality of life.
@JackEllis
@tobias_petry
If you enjoyed this thread you might also enjoy my guide to building a database for data-intensive applications! And if you have any questions please ask! Thanks
@JackEllis
for the prompt :)
This episode with
@kylemathews
is great! I really like his comparison between react and local first. Both allow your app to become more declarative, freeing up cycles to work on your business. Awesome stuff!
Episode
#5
is out.
@kylemathews
on his experience building personal local-first apps with ElectricSQL and how using a data syncing engine makes manual data fetching obsolete.
Just added
@usefathom
to my personal sites using my iPad editing files directly on
@github
auto-deployed by
@Netlify
all in... 5 minutes? Props to
@JackEllis
for a fantastic onboarding experience.
Congrats to
@MindsDB
for raising $7.6M to continue bringing world class ML to the database!! Proud to have helped forge your partnership with
@SingleStoreDB
! Now let's do some data science :)
decided to call the series "local dev pod" and the first conversation-on-air was with
@carlsverre
the author of sqlsync
super fun chat and loved his cloud-as-disk, edge-as-ram analogy
So, last week I upgraded my office with an air purifier + humidifier. This week, I added a CO₂ monitor. Holy crap - with my door and window closed, it hits > 1300ppm which totally explains why I sometimes get headaches. After some optimization it's now hovering around 650 avg.
Anyone using
@obsdmd
? I built a LLM abstraction layer (currently only
@OpenAI
, more coming if requested) designed to be used from other plugins as as plugin! Check it out here: