Sign in

DataFracking

@datafracking.bsky.social
118 followers 202 following 27 posts
PostsRepliesMedia
DataFracking @datafracking.bsky.social · 05/10/2026
here's a combined demo of my Daffy suite of tools, all TDS listeners (think sybase, SQL server, babelfish, type thing) two work with SQL server, one works with duckdb. youtu.be/oML_L6BBOWk?...
youtu.be
Daffy Product Suite Demonstration
YouTube video by Data Fracking
000
DataFracking @datafracking.bsky.social · 03/10/2026
Added DaffyDuo to Daffy Products at TroBeeONE DaffyDB - T-SQL TDS PowerBi capable listener for DuckDB DaffyTee - SQL server Blackbox sql debugging, threat detection, governance, message queue linkage DaffyDuo - SQL server multi-connection management & triage engine www.nuget.org/packages/Daf...
nuget.org
DaffyDuo 0.1.0
A SQL Server gateway that gives one client connection up to three SQL Server connections behind it. Clients connect to DaffyDuo as if it were SQL Server, with a DaffyDuo login; each login starts on it...
001
DataFracking @datafracking.bsky.social · 02/10/2026
youtu.be/ZdPbg2XrrDI?... I have an app that does that too, it's called daffy DB and it does it on DuckDB which can of course then connect to SQL server, postgres SQL, SQL Lite, parquet files, Excel files, etc
youtu.be
DaffyDB 13x faster than SQLServer on a local machine?!?
YouTube video by Data Fracking
000
DataFracking @datafracking.bsky.social · 02/10/2026
you might like this one I made a week ago where I showcase daffy is about 13x (or more)faster than SQL server on a heavy query, direct compare, same T-sql exactly, (credit due that is duckDB doing the heavy lifting) but daffy let's you point a SQL server client, like pbi youtu.be/ZdPbg2XrrDI?...
youtu.be
DaffyDB 13x faster than SQLServer on a local machine?!?
YouTube video by Data Fracking
011
DataFracking @datafracking.bsky.social · 02/10/2026
oh sorry!! yeah I was showcasing my daffydb listener, works on ssms, runs native T-sql, hard to tell it's not SQL server, go 2 even works! it can do aggregate functions that r duckdb native. parquet in example but can connect to SQL server & postgres too @ same time www.nuget.org/packages/Daf...
nuget.org
DaffyDB 0.8.0
Query Parquet, CSV, JSON and Excel files with any SQL Server client. DaffyDB listens on port 1433 and speaks the Tabular Data Stream wire protocol, so Management Studio, sqlcmd, ADO.NET and anything e...
110
DataFracking @datafracking.bsky.social · 02/10/2026
that is actually a very creative answer, but incorrect, I had to look that one up, neat feature! never used it!
110
DataFracking @datafracking.bsky.social · 01/10/2026
absolutely mr context bot, and as you've obviously tried my application you'll realize that it gives you full readout at the end on successes and failures per yaml job entry, so no worries of partials.
000
DataFracking @datafracking.bsky.social · 30/09/2026
I've updated my python based free SQL server ETL tool sqlcarbon on pypi. install with pip. several months since I've changed, new feature target parquet file splits to whatever size you'd like. SQL server to SQL server is mostly unchanged, some new type handling. pypi.org/project/sqlc...
pypi.org
sqlcarbon
Reliable, deterministic SQL Server table-to-table copy tool
100
DataFracking @datafracking.bsky.social · 29/09/2026
s3 example init.sql file for daffy (these also show up as tables in PBI): INSTALL httpfs; LOAD httpfs; CREATE OR REPLACE SECRET s3_config ( TYPE S3, KEY_ID '.', SECRET '.', etc ); CREATE OR REPLACE VIEW v_s3_data AS SELECT * FROM read_parquet('s3://my-bucket/path/to/data.parquet');
000
DataFracking @datafracking.bsky.social · 29/09/2026
director for many years at Dell leading analytics team; realized many users struggle with different file formats & still live in Excel+PBI, plus other use cases w native SQL clients possibles: the daffy->duckDB capabilities also mean native s3, sqlite, postgres etc so a powerful bridge for PBI &more
201
DataFracking @datafracking.bsky.social · 29/09/2026
related to folks in a closed work environment like this, no fabric, and based your depth w/ PBI universe, I'd really be curious what you think of my working product idea, I made it because I had the idea and why not, it's not as simple as it appears, TDS is no joke. youtube.com/shorts/9UAkP...
youtube.com
DaffyDB version 0.7.0 with Power Bi. Index up to 500 csv/excel/json in 2 seconds & auto launch pbi
YouTube video by Data Fracking
110
DataFracking @datafracking.bsky.social · 27/09/2026
old post, I was just searching on Babel fish , not much out there , fyi I did something similar this last month, check out nuget.org and search for daffydb it runs on duckdb using TDS just like Babel but my own way.. and of course duckDB can mount postgres.. so, similar in some ways
000
DataFracking @datafracking.bsky.social · 27/09/2026
daffydb
010
DataFracking @datafracking.bsky.social · 27/09/2026
that would have been a cool thing to add to my demo...
010
DataFracking @datafracking.bsky.social · 27/09/2026
daffydb built on top of duck kinda short circuits some of that.. just sayin... but great article
000
DataFracking @datafracking.bsky.social · 27/09/2026
I made a whole demo turnkey dashboard deal on that a half year back, I assume it still works. You just throw csvs or Excel files at it and it loads it into opfs and duck. I had some other idea on it but kinda fizzled.. trobee.one/ebi/
trobee.one
Excellent BI
000
DataFracking @datafracking.bsky.social · 11/09/2026
politics aside, how is that demoted? Type of episode Rough TV audience Ordinary Kimmel episode 1.4–2.0M Strong guest / interesting news / good lead-in 2–3M Major guest or major event 3–4M+ Exceptional outlier 5–6M+ the episode on YouTube has already gotten over 4 million views.
001
DataFracking @datafracking.bsky.social · 08/09/2026
then also how does the comparison change when you have an A scoring student who is also great with AI. a monster genius? to my earlier example, B kid w great Ai vs A student with no Ai, how does he or she do with orchestration and/or strategic planning tests? vs just standard check the box tests?
000
DataFracking @datafracking.bsky.social · 08/09/2026
I could see that, and it's a concern. no opinion either way, but as a data point I wonder how a mid level scoring kid (not failing medium/B level) that knows how to use AI effectively would fair in more real world tests against the kid with high scores in the specific discipline but no AI skills.
200
DataFracking @datafracking.bsky.social · 29/08/2026
it's fine, just learn how to use a proper sandbox. >sbx run claude literally takes 5 minutes to setup the docker Linux distro for this
000
DataFracking @datafracking.bsky.social · 15/08/2026
Star Trek or the Orville... find out at devildoku com
000
DataFracking @datafracking.bsky.social · 15/08/2026
try my new game devildoku.com
000
DataFracking @datafracking.bsky.social · 30/07/2026
and when they get in, where will those 600k new influx of cars park?? eh?
000
DataFracking @datafracking.bsky.social · 30/07/2026
fun fact, more cyclists equals faster automobile commutes for drivers. If you want to see some major gridlock, force all cyclists to drive in a vehicle on a Monday in New York City. #cycling
110
DataFracking @datafracking.bsky.social · 30/07/2026
have they checked gboard? this is what my pixel is doing today...
000
DataFracking @datafracking.bsky.social · 22/07/2026
yup 👍. it's awesome currently moved to the Qwen 27b MTP, it's closer to 60 t/s same hardware, slower, but it's dense and not a moe, still decently fast and I've found the quality of coding a little better with less bugs both of these models are the best I've used so far for my 32gb vram setup
010
DataFracking @datafracking.bsky.social · 08/07/2026
100+ t/s on the XL, 😎: #llama #vllm llama-server.exe -m "G:\llama\models\qwen3.6-35b-mtp\Qwen3.6-35B-A3B-UD-Q4_K_XL.gguf" --alias "QWEN MTP XL" --spec-type draft-mtp --spec-draft-n-max 3 -c 65536 -ctk q8_0 -ctv q8_0 --kv-unified --ctx-checkpoints 0 -fa on --fit on --fit-target 1536 --jinja -np 1
100