Paul Bruce

📖 Customer study: Deno runs #DenoDeploy on Vultr — achieving global edge performance while reducing infrastructure costs by ~$10,000 per month. 🔗 https://lnkd.in/enNTqK2k Deno Deploy executes #JavaScript and #TypeScript using lightweight V8 isolates at the edge. Every request depends on low latency, efficient routing (#Anycast), and infrastructure that can scale cleanly. By moving from #AWS and #GCP to #VultrBareMetal, #CloudCompute, and #ObjectStorage, Deno was able to maintain its Infrastructure-as-Code model (#Terraform + #container orchestration) without rearchitecting — while improving cost-per-request and edge efficiency. This architecture supports API servers, edge functions, full-stack apps, and secure execution of AI-generated code through #DenoSandbox — with predictable economics as #AI and #LLM workloads grow. Deno is also a #VultrCloudAlliance partner, so teams building modern JavaScript platforms can deploy globally on Vultr using the same production-proven foundation. #CustomerStory #EdgeComputing #CloudInfrastructure Andy Jiang Ryan Dahl

6

In 2017, CreativeGround had outgrown its Drupal 7 architecture. Today, it has a modern UX, API ecosystem, multi-system integration, and 30,000+ profiles. 👀 🌱  It all began with stabilizing an aging platform... and evolved into an 8+ year strategic partnership with the New England Foundation for the Arts (NEFA). ✨ NEFA’s commitment to serving their community aligned perfectly with our focus on human-centered, continuous improvement. 📲 Together, we brought NEFA's ambitious vision to life: A digital infrastructure that connects artists, cultural professionals, and organizations across 6 states. We transitioned the old site to a decoupled Drupal backend with Next.js 14 frontend, creating a solid foundation for:  🔹 Sophisticated search and networking features  🔹 Accessibility and inclusion compliance  🔹 Scalable API architecture   🔹 Seamless multi-system integration Want the full story? Check out the link in the comments! 👇

13

1 Comment

‘[Observability is] one of those things that’s not an optional purchase. You can’t write a whole bunch of software and say, ‘OK, if it goes wrong, I don’t need to investigate that.’ Everybody who writes software, which increasingly is more and more companies, needs to have a strategy for how they’re going to troubleshoot when something goes wrong. So from a channel standpoint, the nice thing is, it’s a must-have product,’ says Observe, Inc. CEO Jeremy Burton.

4

We've added versioning to automations so you can quickly review past configurations and rollback to a previous workflow iteration, and have made some improvements to the visual automation builder. See the details here: https://hubs.ly/Q03LR_vP0

19

Many organizations default to the "known" platform when hosting Drupal — but is that always the best move? Anthony Mangini, our Director of Dev Ops dug into the financial, security, and operational differences between AWS and Acquia for Drupal hosting. The findings challenge some assumptions, particularly for federal agencies and enterprises that prioritize security, compliance, and long-term flexibility. Read the full analysis: https://lnkd.in/gaHBKVey Curious to hear how others are thinking about Drupal hosting in 2025 — feel free to weigh in.

11

"I have only made this letter longer because I have not had the time to make it shorter." - Blaise Pascal'' This also applies to software demos. We've tightened the onboarding workflow and now have a 60 second demo of the journey from a blank slate in Claude Code to a deployed preview application on VeilStream.

2

1 Comment

Next Thursday (Feb 26), Nathan VanBenschoten (Chief Architect at turbopuffer) is joining Doug Turnbull and me (Trey Grainger) for a fascinating Lightning Lesson on how to "𝐒𝐞𝐚𝐫𝐜𝐡𝐢𝐧𝐠 100 𝐛𝐢𝐥𝐥𝐢𝐨𝐧 𝐯𝐞𝐜𝐭𝐨𝐫𝐬 𝐢𝐧 𝐨𝐛𝐣𝐞𝐜𝐭 𝐬𝐭𝐨𝐫𝐚𝐠𝐞". I've worked in information retrieval for decades and with many different open source and commercial search engines and lots of scaling challenges. Usually we have to make significant tradeoffs between query throughput, query latency (milliseconds vs. seconds), number of documents (millions to hundreds of millions), and overall cost (RAM, SSD, distributed clusters, fault tolerance, etc.). Turbopuffer decided to revisit these tradeoffs and build a new search engine "from first principles" with regard to scaling and cost. As the heart of this architecture is a simple idea: a stateless query tier on top of object storage (like AWS S3). This sounds really cool (and inexpensive), but also really slow.... Except somehow turbopuffer has figured out how to do this at scale and still with very fast (sub 200 millisecond p99 latency queries)! Why is this a big deal? It enables nearly infinite scaling (# of docs and queries) and delegates the data distribution layer to VERY CHEAP storage (super cheap), while still maintaining great speed. If you're doing retrieval on a lot of content and concerned about cost and scale, or you're just a search nerd looking to learn more about these scaling dynamics, join us to hear from their Chief Architect about: - Designing a modern search database from first principles - The critical compute vs. bandwidth tradeoffs and bottlenecks - How hierarchical clustering overcomes bandwidth bottlenecks - Binary Quantization at 100 Billion Scale 🔗Sign up link in the comments. Free Lightning Lesson on Maven.

40

1 Comment

Small footprint eXtremeDB offers more features than any other database management system, giving professional developers reliability, speed and connectivity solutions for current needs and future growth.  Unique features include: - Active Replication Fabric to mitigate IoT connectivity issues - A *hard* real-time option - In database analytics. Pipeline over 150 built-in vector based statistical functions for the lowest possible latency when processing time series data Learn more: https://t.ly/y0nf- #EmbeddedDevelopment #extremedb #DatabaseManagement for #embeddedsystems #iotsolutions #indatabaseanalytics

7

Event noise is the #1 reason ITOM rollouts stall. Because every team drowns in thousands of alerts, and no one knows which ones matter. ServiceNow ITOM solves this with a 3-Layer Model that separates noise from signal: Layer 1 → Noise Reduction ✔️ Suppress duplicates. ✔️ Filter non-actionable alerts. Layer 2 → Event Correlation ✔️ Merge related alerts into service-level events. ✔️ Stop firefighting every single ping. Layer 3 → Business Impact Mapping ✔️ Tie events to CIs and services. ✔️ Escalate only when revenue, users, or customers are at risk. This model transforms your NOC from reactive firefighting → proactive decision-making. And it shows executives what they actually want: Not “alerts closed.” But “business protected.” Follow for more: https://lnkd.in/eM6bbun9 #ServiceNow #ITOM #EventManagement #Observability #ServiceMapping #ServiceNowConsulting

1

2 Comments