Blosc Development Team
LightNews LightNews
blosc.org
Blosc Development Team
@blosc.org
28 followers 4 following 40 posts
Announcements about Blosc2 developments
https://blosc.org
Posts Replies Media Videos
blosc.org
🚀 Boost your Python computations on NumPy arrays with @blosc.jit! Let Blosc2's compute engine accelerate your code without changing containers. ⚡

For ultimate speed, native Blosc2 containers are still king. 🏎️

Code: gist.github.com/FrancescAlte...

#Python #DataScience #NumPy #Performance #Blosc
September 22, 2025 at 12:19 PM
blosc.org
🚀 We're thrilled to announce *TreeStore*, a new class in Python-Blosc2! Endow your datasets with a hierarchical structure! ⚡️

📝 We've blogged about it:
www.blosc.org/posts/new-tr...

It's in beta, and available in Python-Blosc2 v3.7.2. Enjoy!

#Python #Blosc2 #TreeStore #DataScience #OpenSource
August 20, 2025 at 6:07 AM
blosc.org
Python-Blosc2 comes with an extensively tested partition sizes algorithm for automatically setting the best chunk and block sizes for you.

But you can always bypass this mechanism for more appropriate fine-tuning in different scenarios 🚀

www.blosc.org/python-blosc...

#Compression #Performance
July 29, 2025 at 8:43 AM
blosc.org
🗣️ Announcing Python-Blosc2 3.6.1

¡Unlock new levels of data manipulation with #Blosc2! 🚀
We've tamed the complexity of fancy indexing to make it intuitive, efficient, and consistent with NumPy's behavior.

👉 www.blosc.org/posts/blosc2...

Enjoy

#Python #DataScience #BigData #NumPy #Performance
New Blosc2 fancy indexing performance against other competitors, including NumPy. Blosc2 is quite efficient.
July 18, 2025 at 9:31 AM
blosc.org
🚀 Blosc2 supports memory-mapped files for super-efficient data access! 🚀

✨ Why memory-mapping?
1️⃣ No system call overhead for each read/write
2️⃣ Data goes straight from page cache to user space—much faster than traditional I/O!

👉 github.com/Blosc/python...

#DataScience #Performance #BigData 🚀💾
July 14, 2025 at 5:18 PM
blosc.org
#Blosc2 now runs directly in your browser! Leveraging the power of #WASM, #Pyodide, and #JupyterLite, you can harness efficient, adaptable compression through the web's universal interface.

Compress Better, Compute Bigger, Share Faster

#WebAssembly #DataCompression #WebDevelopment #DataScience
July 8, 2025 at 10:53 AM
blosc.org
🗣️ Announcing Python-Blosc2 3.5.1 🚀

This introduces significant performance and memory optimizations, enhancing the experience of computing with large, compressed datasets using lazy expressions.

Compress Better, Compute Bigger!

#Performance #MemoryEfficiency #DataScience
Evaluate slices of lazy expressions (inner dims)
July 2, 2025 at 5:43 PM
blosc.org
Struggling with slow I/O in #HDF5? Try #Blosc2 as a filter or I/O data handler — faster data, less pain!

👉 www.blosc.org/posts/pytabl...

#Performance #DataScience
July 1, 2025 at 12:24 PM
blosc.org
📣Python-Blosc2 3.5.0 is out! 🚀

This release introduces the powerful stack() function! 🥞 You can now stack multiple Blosc2 NDarrays along a new axis, similar to np.stack(), while getting great performance.

We have blogged about it here: www.blosc.org/posts/blosc2...

Compress Better, Compute Bigger
June 24, 2025 at 3:52 PM
blosc.org
📢 We are pleased to announce the integration of a new stack feature in #Blosc2 🚀, which allows for stacking large arrays along a new axis.

We've updated our recent blog post:
Check it out! 👇
www.blosc.org/posts/blosc2...

Compress Better, Compute Bigger
#Python #DataScience #Performance #OpenSource
June 23, 2025 at 4:34 PM
blosc.org
#Python-Blosc2 is hitting 1 million weekly downloads on PyPI! 🎉

Users are rapidly adopting #Blosc2, which now accounts for over 95% of downloads compared to Blosc1. 📈 Thanks to our amazing users. 🙏 🚀

Our motto: Compress Better, Compute Bigger! 💪

#Milestone #CommunitySupport #DataCompression
June 5, 2025 at 4:24 PM
blosc.org
💡 Did you know you can supercharge your #HDF5 datasets with #Blosc2? 🚀

Leverage hdf5plugin (hdf5plugin.readthedocs.io) to integrate Blosc2 as a filter within HDF5. Create and read data using popular Python wrappers like h5py or PyTables, while achieving excellent performance! 💨

Compress Better!
May 20, 2025 at 8:25 AM
blosc.org
🗣️New blog entry of Ricardo Sales Piquer, our intern working in linear algebra problems:

Make NDArrays Transposition Fast (and Compressed!) in #Blosc2 🚀

www.blosc.org/posts/optimi...

Great work Ricardo! 🎉

#Compression #LinearAlgebra #Optimization #DataScience

Compress Better, Compute Faster 😀
Performance of Blosc2 transposition depending on the chunk size
April 30, 2025 at 10:37 AM
blosc.org
📣 Announcing the release of Python-Blosc2 3.3.0:

🔄 New blosc2.transpose()
⚡️ New fast path for NDArray.slice() - up to 40x speedup
🔧 NDArray.slice() now preserves compression params
📝 Improved documentation

Thanks to all contributors!

github.com/Blosc/python...

Compress Better, Compute Bigger! 😀
Transposing a compressed array with Blosc2
April 9, 2025 at 4:28 PM
blosc.org
📢 🔥 Article on Blosc2: Compute with TB-sized datasets on your own hardware!

🚀 Outperforms NumPy by 10x ~ 100x for large computations
🐍 Integrates seamlessly with the Python data science ecosystem
💻 Works both in-memory and on-disk with minimal performance differences

ironarray.io/blog/compute...
Up to 75x speedups of Blosc2 vs NumPy and Numba when computing with generic NumPy functions
March 28, 2025 at 1:48 PM
blosc.org
🔥 Blosc2 3.2.1: NumPy's best friend! 🔥 blosc2.jit + NumPy array interface = blazing fast, direct compressed data crunching!

See how: github.com/Blosc/python-blosc2/blob/main/examples/ndarray/jit-numpy-funcs.py

#DataScience #Python #Compression

Compress Better, Compute Bigger!
The plot shows the performance of the new Blosc2 compute capabilities. It can be more than 10x faster than NumPy and numba.
March 26, 2025 at 6:04 PM
blosc.org
Did you know that you can use the @blosc2.jit decorator to accelerate your computations with NumPy by 10x and more by just adding a single line to your function?

No need to add loops manually, just add @blosc2.jit and it will do its magic!

Compress Better, Compute Bigger!
blosc2.jit can accelerate computations a lot, and very easily.
March 7, 2025 at 1:56 PM
blosc.org
Blosc compression helped fighting memory bottlenecks for 15 years now! 🎂
See how Dask + Zarr benefits from it. But vertical integration between compression and the compute engine in newest Python-Blosc2 makes a big difference in terms of speed ⚡ and scalability 🤟

#Compress better, #Compute bigger!
Performance of Blosc2 vs Dask+Zarr+Blosc for a complex math calculation.
February 25, 2025 at 4:45 PM
blosc.org
Did you know that with python-blosc2 you can select an arbitrary precision of floating point data? This allows for a much better compression ratio: from 2x up to 100x. And still performing like a champ ⚡⚡

More info: www.blosc.org/python-blosc2/

Compress better, compute bigger!
#compute #compress
Compute performance vs arbitrary precision
February 19, 2025 at 6:31 PM
blosc.org
📢 For Python-Blosc2 3.1.1 we have optimized even more the shiny new compute engine.

See how the new version can reach great performance even for the case that the operands largely exceed (by up to 20x) the RAM of your computer ⚡️

www.blosc.org/python-blosc2/

Compress better, compute bigger!
Compute performance scales well beyond the available RAM
February 14, 2025 at 1:17 PM
blosc.org
🗣️ C-Blosc2 2.16.0 is out. We have added new features, like:

* _fseeki64/_ftelli64/_stat64 on Windows for large file support.
* 12-byte unshuffle for avx2 and sse2.
* Better description of the Blosc2 format as a whole.

Compress better, Compute faster!
Double partitioning diagram
February 12, 2025 at 10:43 AM
blosc.org
Did you know that:

$ conda -c conda-forge install python-blosc2 mkl

will let you fully utilize your CPUs?

⏩ Advanced SIMD
⏩ Multiple cores
⏩ Leverage caches efficiently

www.blosc.org/python-blosc...

Do not let your hardware sit idle; put it to work for you. Kudos to @conda-forge.org team! 🤟
Compare performance of Blosc2 against NumPy, Numexpr, Numba and Dask.
December 19, 2024 at 5:31 PM
blosc.org
📢 The Blosc Development Team is proud to announce the release of Python-Blosc2 3.0.0 final! 🎉 🎉

See an introduction to the shiny new compute engine at: www.blosc.org/python-blosc...

Looking forward to what you can build with Python-Blosc2 3.0!

** Compress Better, Compute Bigger! **
Compare reductions on different axis of a 4D array, using NumPy and Python-Blosc2
December 13, 2024 at 12:47 PM
blosc.org
📢 We are adding the latest bolts and nuts to the Blosc2 library before release 3.0.0.
In our last release candidate, we fixed a couple of issues that showed up in our recent tutorial at the PyData Global 2024.

github.com/Blosc/python...

We are curious on what you are going to build with it! 😀
Comparison of computation on compressed vs uncompressed ndarrays
December 11, 2024 at 1:28 PM