iiinova — where your digital self keeps evolving

Joe Hamman @jhamman.bsky.social · 22/01/2026

We watched data teams waste years building the same ingestion pipelines for weather data. So we fixed it. Zero ingestion. High velocity. Cloud-native. Proud to launch with incredible partners like @brightband.com, #zeusai, and dynamical.org. Give it a spin! 👇

120

Reposted by Joe Hamman

Earthmover @earthmover.io · 04/11/2025

Wildfire season may be abating, but our need to understand future scenarios has not: Come join CTO @jhamman.bsky.social Thursday morning for a workshop: Predicting cloud native wildfire risk with the Earthmover platform: m.earthmover.io/predictingwi...

011

Reposted by Joe Hamman

Earthmover @earthmover.io · 23/09/2025

Meet @rabernat.bsky.social, @jhamman.bsky.social and more of the Earthmover team at our #ClimateWeekNYC events. PSA: the hands-on tech workshop Wednesday and panel on Thursday still have capacity, but are filling up fast.

021

Joe Hamman @jhamman.bsky.social · 10/07/2025

Super excited to see #icechunk v1.0 ship today. Stable format, stable API, and ready for production. Take it for a spin and let us know how it goes. 🚀🚀🚀

070

Joe Hamman @jhamman.bsky.social · 07/05/2025

Just back from #CNG2025, I wrote up some thoughts on the traction the @zarr.dev format is gaining in the broader geospatial industry (specifically Earth Observation). I also summarize what features and tools the user and developer communities are asking for to fully leverage Zarr. 👇👇👇

042

Joe Hamman @jhamman.bsky.social · 28/04/2025

I'll be at the CNG conference in Snowbird next week. I wrote a short blog post about what the Earthmover team will be up to. tldr; we'll be talking about @zarr.dev, #icechunk, @xarray.bsky.social and cloud-native data cubes. Details in the blog post 👇

020

Joe Hamman @jhamman.bsky.social · 24/04/2025

Most people think of @zarr.dev as a "file format". With #Icechunk, we've turned Zarr into a database. @functionth.bsky.social's post shows how Icechunk can be used to solve a problem where transactional databases are often required.

010

Reposted by Joe Hamman

Earthmover @earthmover.io · 23/04/2025

1/ 🚨 New Blog Post Alert: "𝐿𝑒𝑎𝑟𝑛𝑖𝑛𝑔 𝐴𝑏𝑜𝑢𝑡 𝐼𝑐𝑒𝑐ℎ𝑢𝑛𝑘 𝐶𝑜𝑛𝑠𝑖𝑠𝑡𝑒𝑛𝑐𝑦 𝑤𝑖𝑡ℎ 𝑎 𝐶𝑙𝑖𝑐ℎ𝑒́𝑑 𝑏𝑢𝑡 𝐼𝑛𝑠𝑡𝑟𝑢𝑐𝑡𝑖𝑣𝑒 𝐸𝑥𝑎𝑚𝑝𝑙𝑒" 🏦🔁 👉 Read it here: earthmover.io/blog/learnin...

earthmover.io

Learning about Icechunk consistency with a clichéd but instructive example - Earthmover

In this post we’ll show what can happen when more than one process write to the same Icechunk repository concurrently, and how Icechunk uses transactions and conflict resolution to guarantee consisten...

151

Joe Hamman @jhamman.bsky.social · 21/04/2025

There's still time to sign up for tomorrow's webinar on ⚡Flux ⚡ 👇 share.hsforms.com/1FLrE9CpwSLq...

share.hsforms.com

Form

000

Joe Hamman @jhamman.bsky.social · 21/04/2025

@zarr.dev and #icechunk are amazing but they are not magic. They are part of a thoughtfully designed cloud-native data architecture. @tegnicholas.bsky.social peels back the covers on cloud-optimized scientific data formats in our latest "Fundamentals" post 👇

021

Joe Hamman @jhamman.bsky.social · 21/04/2025

We found similar results when we first benchmarked #icechunk. Our conclusion: doing IO with a Rust backend is much faster than Python. 👇Really exciting to see @kylebarron.dev's Obstore backend for Zarr-Python ship today.

0112

Joe Hamman @jhamman.bsky.social · 16/04/2025

No more bespoke APIs. No glue code. Just fast, interoperable access via tools you already use — QGIS, ArcGIS, web maps, Python, R, MATLAB, and more. Flux is a cheat code for building data products.

100

Joe Hamman @jhamman.bsky.social · 16/04/2025

Flux serves data directly from Icechunk via standard APIs like EDR, WMS, and OPeNDAP. It’s stateless, scalable, and cloud-native. From raw data to production-grade API endpoints in seconds.

100

Joe Hamman @jhamman.bsky.social · 16/04/2025

We built Flux to solve a core pain point: turning massive geospatial datasets into usable APIs was slow, brittle, and distracting. Flux makes it trivial for any data producer to become a data provider.

100

Joe Hamman @jhamman.bsky.social · 16/04/2025

🚀 Big launch today: introducing Flux — a game-changing way to serve geospatial data via standard APIs in seconds. Built for scale, speed, and simplicity.

153

Joe Hamman @jhamman.bsky.social · 09/04/2025

Training AI models at scale from data stored in cloud object storage requires thinking carefully about both bandwidth and concurrency. In this post, @functionth.bsky.social get’s into the details of concurrent reads at scale, showing how #Icechunk and S3 can easily scale beyond 200k requests/second!

120

Reposted by Joe Hamman

Earthmover @earthmover.io · 09/04/2025

📣 Blog post alert! 𝐄𝐱𝐩𝐥𝐨𝐫𝐢𝐧𝐠 𝐈𝐜𝐞𝐜𝐡𝐮𝐧𝐤 𝐬𝐜𝐚𝐥𝐚𝐛𝐢𝐥𝐢𝐭𝐲: 𝐮𝐧𝐭𝐚𝐧𝐠𝐥𝐢𝐧𝐠 𝐒𝟑'𝐬 𝐩𝐫𝐞𝐟𝐢𝐱 𝐬𝐭𝐨𝐫𝐲. This technical post by @functionth.bsky.social dives deep into the internals of how S3 shards data, showing that distributed Icechunk can easily perform 230,000 object reads/sec and beyond. earthmover.io/blog/explori...

earthmover.io

Exploring Icechunk scalability: untangling S3's prefix story | Earthmover

We show Icechunk can scale to extremely high concurrency levels, and explain how it achieves this in modern object stores.

254

Joe Hamman @jhamman.bsky.social · 03/04/2025

We often see folks try to convince tabular data tools to perform well with multi-dimensional array data. This post by @rabernat.bsky.social explains, from first principles, why this rarely works. Its a good one! 👇👇👇

131

Joe Hamman @jhamman.bsky.social · 28/03/2025

🚨 New blog post 🚨 In it, we show off our recent work deploying #icechunk on top of #NASA's existing archives of Earth observation data. The results: 100x speed up when extracting time series from existing datasets stored as netCDF.

0114

Joe Hamman @jhamman.bsky.social · 25/03/2025

This session is going to be a blast! If you are headed to CNG next month (and you should be!), consider joining us for this workshop on @xarray.bsky.social , @zarr.dev , and #icechunk. 👇👇👇

110

Reposted by Joe Hamman

Earthmover @earthmover.io · 18/03/2025

1/ As a data platform company, we get a unique view of how other companies use data. In light of the recent cuts at @noaa.gov, our CEO, @rabernat.bsky.social, reached out to our customers to get their perspective on the value of NOAA’s data and services.

1104

Joe Hamman @jhamman.bsky.social · 05/03/2025

I'll be giving a short update tomorrow on the state of Zarr-Python following the 3.0 release back in January. Swing by the Pangeo showcase to catch the latest progress. @zarr.dev

060

Reposted by Joe Hamman

Earthmover @earthmover.io · 27/02/2025

Denver folks — join us for a Geospatial Happy Hour on Wednesday, March 12th at 7 PM MST. Come enjoy drinks, tacos, discussing geospatial data, and meeting the Earthmover team. Register: lu.ma/o6r39p6j

lu.ma

Denver Geospatial Happy Hour · Luma

Come have a drink, grab a bite, and nerd out with the Earthmover team on geospatial data. Earthmover will provide food and drink tickets. Our event will be…

053

Reposted by Joe Hamman

TEGNicholas.bsky.social @tegnicholas.bsky.social · 27/02/2025

We now have a public slack channel for VirtualiZarr development and questions, feel free to join! earthmover-community.slack.com/archives/C08...

earthmover-community.slack.com

Slack

054

Joe Hamman @jhamman.bsky.social · 27/02/2025

Very, very stoked to have @tegnicholas.bsky.social join the Earthmover team!

150

Joe Hamman @jhamman.bsky.social · 24/02/2025

I'm curious what sort of issues you are seeing. I'm able to open datasets on s3 without any problems.

200

Joe Hamman @jhamman.bsky.social · 24/02/2025

I rode the race with my good friend Nico! Together, we rode about 320 miles over 24 hours! www.strava.com/activities/1...

030

Joe Hamman @jhamman.bsky.social · 24/02/2025

Here on Bluesky, I follow two topics: 1) software & data infrastructure and 2) cycling. I mostly post about software but today, I can share a bit about a fun cycling adventure of my own 🚵‍♂️. Last week, I got to ride in the 24 Hours in the Old Pueblo - placing 4th in the Open Men Duo category:

2141

Joe Hamman @jhamman.bsky.social · 24/02/2025

Zarr-Python 3 included a bunch of performance improvements that are now landing in @xarray.bsky.social. Check out @davisvbennett.bsky.social's post on what we changed in Xarray/Zarr and how it is speeding up the "time to first byte" for large datasets 👇

040

Joe Hamman @jhamman.bsky.social · 17/01/2025

If you missed the Zarr-Python 3 release party last week, no worries because today, we released zarr-python 3.0.1 and numcodecs 0.15.0. release notes 👉 zarr.readthedocs.io/en/stable/re...

zarr.readthedocs.io

Release notes — zarr 3.0.1 documentation

0102

Joe Hamman @jhamman.bsky.social · 10/01/2025

Shards are the unit of writing so the upper limit is really what you can fit in memory at once (yes some stores will support partial writes).

110

Joe Hamman @jhamman.bsky.social · 10/01/2025

It's going to depend on a number of things and we likely have some benchmarking and tuning to do, but I have a hunch that we'll start to see chunks <1mb and shards >1gb.

230

Joe Hamman @jhamman.bsky.social · 09/01/2025

Come to our booth at AMS where we'll be giving out demos of Arraylake and Icechunk as well as Rubik's cube lessons!

010

Reposted by Joe Hamman

Earthmover @earthmover.io · 09/01/2025

We’re thrilled to announce the release of @zarr.dev-Python 3! This major release brings full support for Zarr v3 specification, including the new chunk-sharding extension, major performance enhancements, and a thoroughly modernized codebase. Read more: earthmover.io/blog/zarr3-r...

earthmover.io

Zarr-Python 3 is here! | Earthmover

Zarr-Python 3 is here! This release brings support for Zarr's v3 specification, new extensions, and major performance improvements.

082

Reposted by Joe Hamman

xarray.bsky.social @xarray.bsky.social · 09/01/2025

Our friends over at @zarr.dev made a big release today. Xarray v2025.01.1 was also released today with support for Zarr-Python 3 🚀

0101

Reposted by Joe Hamman

Zarr @zarr.dev · 09/01/2025

🎉 Zarr-Python 3 is here! 🎉 - Full support for Zarr v3 spec - Chunk-sharding for more efficient data storage - Major performance boosts with async I/O & parallel compression 💻 pip install --upgrade zarr 💻 conda install --channel conda-forge zarr Blog post: buff.ly/3C3OwYw

25520

Joe Hamman @jhamman.bsky.social · 09/01/2025

The 3.0.0 release clears the way for a bunch of exciting extensions built on top of the v3 spec. #icechunk, variable chunking, new dtypes, and more are all now possible. Time to get busy.

000

Joe Hamman @jhamman.bsky.social · 09/01/2025

One of my favorite parts of this release is that it included contributions from over 30 individuals. After a period of slow development, the version 3 refactor really managed to build momentum in the project. Special shout out to @normanrz.com and Davis Bennett for doing a big chunk of the work!

120

Joe Hamman @jhamman.bsky.social · 09/01/2025

We haven't done much performance tuning but the early results are looking really good.

performance comparison between zarr 2 and zarr 3

101

Joe Hamman @jhamman.bsky.social · 09/01/2025

We released Zarr-Python 3 today, complete with support for the v3 spec, new extensions (chunk-sharding) and significant performance gains. earthmover.io/blog/zarr3-r...

earthmover.io

Zarr-Python 3 is here! | Earthmover

Zarr-Python 3 is here! This release brings support for Zarr's v3 specification, new extensions, and major performance improvements.

1134

Reposted by Joe Hamman

Zarr @zarr.dev · 04/01/2025

Hi, Zarr community! 👋 🚨 Big news: Zarr-Python 3.0 is dropping next week! 🚨 If your project depends on Zarr-Python, make sure to check out the migration guide to get ready for the update. Guide: buff.ly/3W0cLxd More updates soon! 🧑🏻‍💻

buff.ly

02713

Joe Hamman @jhamman.bsky.social · 18/12/2024

I'm still recovering from #AGU24 but I'm already thinking ahead to #AMS25. Who's going to be in New Orleans and who wants to meet up?

290

Joe Hamman @jhamman.bsky.social · 16/12/2024

Final point here. The architecture here is quite similar to many of the big tabular data systems out there today. Snowflake, Big Query, Databricks all have converged on a similar model. Evidence that the architecture works for a broad set of applications and users.

020