David Ingham

🚀 Want to run your very own trained LLM, privately and securely? 🤖🔒 Imagine harnessing the power of a Language Model tailored just for you without worrying about third-party servers! 💼✨ Try, LM Studio. Jest for few click . LLM running in your PC. With LM Studio, you can ... 🤖 • Run LLMs on your laptop, entirely offline 📚 • Chat with your local documents (new in 0.3) 👾 • Use models through the in-app Chat UI or an OpenAI compatible local server 📂 • Download any compatible model files from Hugging Face 🤗 repositories 🔭 • Discover new & noteworthy LLMs right inside the app's Discover page #LLM #private #chatgpt #AI

9

This is a great little video about the dangers of serverless cloud computing and ensuring your side project is monetised before it goes viral. https://lnkd.in/e8KxSb5U

7

2 Comments

Lots of interest in setting up Observability with open source tools today at my Unconference session today @opentelemetry is a big part of how people are thinking of re imagining their Observability stacks in 2024 and cost is major consideration #OSSummit

95

12 Comments

here is an excellent article on the role of the developer in the age of AI https://lnkd.in/eTkvrkBQ

12

With the announcement of S3-native-streams (Freight clusters), here is a commentary on Confluent strategy regarding object storage, streaming and an open data architecture. https://lnkd.in/dTg6DKVy

49

6 Comments

“If you can't explain it simply, you don't understand it well enough” - Albert Einstein Our mission: 🚀 “Backplane’s strategy is really simple. We want to put all the incredibly great cloud platforms into a single API that will simplify and democratise multicloud consumption to enable every organisation in the world to obtain the best value pricing from Cloud.” Our vision: 🌍 Build the global entry point to cloud Our values: ❤️ People, Purpose, Passion. It’s that simple! To find out more, try out the open source cloud abstraction API here: https://getbackplane.io

15

You too can follow along with my new article and load 1 TiB of data / 2 billion rows into a SingleStore Trial cluster in under 20 minutes. In the case of data loads, impatience is a virtue. https://lnkd.in/grFBJvrD

45

4 Comments

Thanks, Paul Vincent Modernization has become a top topic of Gartner inquiry. And to modernize, it helps to know what modern is. So, “Future of XYZ” is a popular topic these days. https://lnkd.in/eBmBrV6J #change #leadership #future #risk #modernization #transformation #readiness #preparedness #composable #leadership #mindset

5

Easy as Pie: Stateful Services at CloudKitchens https://lnkd.in/eDDTfs5e

3

What have we learned from more than a decade of deploying microservices? And are we any better at breaking large systems down into the optimum slices? Ian Cooper and James Lewis join me to thrash it out, in this live episode recorded at GOTO Copenhagen…

37

2 Comments

Read my latest covering Syntasso’s Daniel giving names to the three ways to build an #IDP, only on The New Stack.

8

If you have been considering migrating a Teradata, Oracle or SQL Server database to Snowflake, good news! We've released a free version of our SnowConvert toolkit that can convert table DDL and offer code assessments on scripts and stored procedures. This can cut large amounts of time out of your migration. You can find more details on this blog post https://lnkd.in/gmV_Q3qF #snowflake #snowconvert #datamigration #oracle #sqlserver #teradata

10

The SCVO guides are always super clear. Some great practical advice in the latest one on low-cost digital

2

Incredibly proud of the latest Kinesis Client Library version (3.0) which, when compared to previous KCL versions, reduces compute costs by up to 33%. KCL 3.0 introduces an enhanced load balancing algorithm that continuously monitors resource utilization of the stream processing workers and automatically redistributes the load from over-utilized workers to other underutilized workers. This ensures even CPU utilization across workers and removes the need to over-provision the stream processing compute workers which reduces cost. Additionally, KCL 3.0 is built with the AWS SDK for Java 2.x for improved performance and security features, fully removing the dependency on the AWS SDK for Java 1.x. #Kinesis #DataStreaming #KCL

94

Is storage really so cheap? It might not seem so when you look at the ever-growing cloud bill. The price of S3, for example, hasn’t really changed in the last 8 years, and in my view, it's more likely to increase rather than decrease. Elon’s reposted chart on falling storage hardware costs (HDDs, SSDs, flash) is fascinating, but it doesn’t tell the full story. While hardware costs are dropping, the total cost of ownership (TCO) for storage often increases due to modern demands: high performance, scalability, security, compliance, and managing massive data volumes. Cloud storage costs can also include network transfer, data management, and protection overhead. Cheaper hardware doesn’t always mean cheaper storage operations—scale and complexity often outpace those savings. The software stack also needs to catch up, as it must efficiently manage the growing complexity of both data volume and technological demands. And, if there are any savings left, they’re typically to the benefit of the cloud provider rather than the customer. Thoughts?

29

4 Comments

For those of you still on the fence about OpenTofu, it's time to get off the fence. I used to have to dance around swapping provider configs in terraform to test my code against #localstack. Dynamic providers should have been built by Hashicorp forever ago. It would have made the creation of well tested pipelines so much easier. We finally have a team of contributors building the practical features operators need instead of hiding them behind a paywall. If you are using a pre-BUSL version (and I know you are), swap your binary to tofu on one pipeline at a time if you want to be cautious and you are done. The real magic is in the provider registry which can't change because of the number of people who do not work for Hashicorp who would have to update their provider. The risk is lower than you think it is.

142

12 Comments

Nice to see Cloudsmith listed as the #1 fastest-growing tool in this year's CircleCI State of Software Delivery report - https://lnkd.in/ejnSYr62. If you're building software with CircleCI, you should consider pointing CircleCI at Cloudsmith as your artifact repository (for lots of reasons). Try it out at https://try.cloudsmith.com.

125

2 Comments