
CTO@Yellowbrick. Redditch * Brixham * Leeds #lufc * Newcastle * Malvern * San Francisco * San Diego * Raleigh
Raleigh, NC
Joined March 2009
- Tweets1K
- Following323
- Followers418
- Likes339
This is a wonderful paper: arxiv.org/pdf/quant-ph/01010….
Thanks to @skdh for posting a link. The removal in the 5th axiom of the word “continuous” gives classical probability theory; leave it in, you get quantum probabilities with complex amplitudes.
I've written a blog on how Floecat can generate NDV stats via sampling. NDV (# of distinct values) is one of the most important stats for SQL planning: eng-floe.github.io/floecat/2…
Floecat is an open source Iceberg and Delta table catalog of catalogs that can generate statistics.
My colleague at FloeDB has a new blog explaining the virtues of the VARIANT data type, and its future adoption in data lakehouses: floedb.ai/blog/how-variant-w…
I was at the Iceberg Summit 2 weeks ago and in the final panel it was suggested that VARIANT is an under-hyped type. I agree
Thanks for inviting us to present!
The founders of FloeDB (@markcusack + Kurt Westerfeld) gave an interesting talk with @CMUDB about their new @ApacheIceberg-compatible query engine. Two key takeaways from their talk:
1⃣ Floe is a hard fork of @YellowbrickData.
2⃣ Floe is building a "catalog-of-catalogs"
youtube.com/watch?v=Kq3csHJq…
Lads, it’s been 45 years since we last beat Manchester United Football Club at Old Trafford. 1981. I’d just taken possession of my first digital watch at the age of 11. The space shuttle first launched 6 weeks after the win, the birth of the internet passed by. And here we are 😂
I’ve never waited so long to see a porch erected. I’m expecting something very grand. A vestibule, if you will. #artemis
Mark Cusack retweeted
After 30+ years, "Is Linux disk I/O finally fast enough?" floedb.ai/blog/after-30-year…
Daniel Farke’s tactical plan be this #LUFC
youtu.be/Cgy6Fcz7q0s?si=lBkS…
Mark Cusack retweeted
Streaming inputs and Data Warehouses create RAG based LLMs on steroids. An interesting conversation just had with @markcusack from @YellowbrickData on the @GTwGTPodcast
Dropping late November...
Mark Cusack retweeted
Explore the pros and cons of SaaS-based data platforms. Check out this article from Yellowbrick CTO @MarkCusack ahead of #KubeCon + #CloudNativeCon on @TheNewStack now! bit.ly/3ZO1XoU
Mark Cusack retweeted
Enterprise Data Platforms on Kubernetes Challenge Status Quo | By @markcusack of @YellowbrickData, thanks to @CloudNativeFdn
#Kubernetes #Database
thenewstack.io/enterprise-da…
Got my daft US kettle on at 120v in the hope it boils before this baby arrives @TheSquareBall. What a day.
Mark Cusack retweeted
Check out how @NCSolutions leverages Yellowbrick's Data Warehouse and Snowflake Marketplace to gain real-time insights into ad performance. Learn how you can use both solutions to empower your customers.
#Yellowbrick #SnowflakeMarketplace #BetterTogether okt.to/hCAkms
Mark Cusack retweeted
Feels weird to just go out and buy someone who seems obviously well qualified. Not very Leedsy. I like it.
I can confirm the confirmation of a rumour that #TaylorSwift will be performing at the Chelsea Wrexham game tomorrow. Massive if true. I’ll be there. Stand by for updates. instagram.com/p/CrthvgTvQwN/…
Mark Cusack retweeted
We know you are feeling the pain of #datawarehousing cloud costs.
On Feb. 23rd @DMRadioOnline segment Analyst @YvesMulkers of @7wData and #Yellowbrick CTO @MarkCusack will reveal cloud warehouse savings due to the underlying architecture.
Register below.
Mark Cusack retweeted
News: Yellowbrick has been named a leader in #DataWarehousing by @Gigaom!🏆
We are being recognized for advantages in
💸Cloud cost
💨Speed
🙌 Scalability
🟨And more
Check out our CTO @markcusack’s blog to learn how we went from newcomer ➡️ contender.
okt.to/UbtMhJ
Slow Emotion Replay.
The 30th anniversary of Dusk. Great memories of one of my favourite albums and great times with @thethe













