Loading…
Topic
160 posts about Deployment. Every summary links to the original.
See the idea
A/B testing trades traffic, time, and exposure risk for evidence. This explainer maps the benefit of a controlled comparison, the real cost of running one, and the condition that decides when measuring beats shipping on judgment.
Bun packs a runtime, package manager, bundler, and test runner into one binary. Here is how those four parts fit together and where the seams still show.
CloudFormation is a declarative reconciler: you describe the target state in a template, and it provisions, updates, and reports on the stack that results. Here is the loop.
The Good Stuff
Paid picks we use ourselves. Links carry a referral code — a subscription through one earns us a commission at no extra cost to you.
Newest first
Supabase’s January 2023 beta update gathers releases across Postgres, GraphQL, Edge Functions, client libraries, image transformation, and self-hosting. The update adds pgvector, a PostgreSQL extension for storing OpenAI embeddings and performing vector similarity search, and presents Supabase Clippy, a one-week, fully open-source documentation-search MVP combining pgvector with OpenAI. pg_graphql now supports Views, Materialized Views, and Foreign Tables, while Image Transformations enable automatic WebP conversion by default, with format: origin available to opt out. Other changes include Python and C# reference documentation, pg_repack, multi-tab authentication logout through BroadcastChannel, lz4 compression, ICU locales, downloadable Edge Function source, and a DigitalOcean deployment guide using Sendgrid, Packer, and Terraform. Supabase also reports reaching 45,000 GitHub stars and a position among the top 200 most-popular repositories.
Ruby 3.2 makes YJIT, Shopify’s just-in-time compiler for CRuby, substantially more suitable for production after its experimental Ruby 3.1 release. The project was ported from C99 to Rust for stronger safety guarantees and better abstractions, while memory work reduced overhead to approximately one third of the Ruby 3.1 level through metadata optimization, machine-code garbage collection, and lazy page allocation. YJIT 3.2 speeds railsbench by about 38% over the Ruby 3.2 interpreter, or 57% over the Ruby 3.1.3 interpreter, and adds native ARM64 support for Apple M1/M2, AWS Graviton, and Raspberry Pi. Shopify has deployed it across its Storefront Renderer infrastructure, measuring real-world speedups of 5% to 10%, while identifying memory footprint and method-call optimization as continuing priorities.
Nextdoor deploys weekly mobile updates across iOS and Android using phased rollouts starting at 1% adoption to minimize blast radius. Standard aggregate observability cannot reliably detect early regressions because early adopters skew significantly more active than average users, obscuring silent drops in overall metric noise. To overcome this selection bias, Nextdoor built App Release Anomaly Detection using difference-in-differences causal inference. The methodology verifies pre-adoption trends with standard deviation bounds and fits a linear regression model to estimate the effect against an unobserved counterfactual. During a rollout of iOS version v1.234.5, this model identified a statistically significant app session decline at 1% adoption, isolating the regression 10 days earlier than week-over-week metrics.
Supabase announced the winners of its sixth hackathon, held during Launch Week 6 after ending 2022 with many feature releases. Projects were judged across categories including Best Overall, Most Fun/Interesting, Best Flutter, Best Storage, and Best Edge Functions, with Deno participating in judging the edge category. Depulso, a CLI and web app for static site deployments with per-deployment .depulso.site subdomains and SSL, won Best Overall, while Supa SQL App was runner-up for converting human language into SQL queries. Other winners included HTTP Sense, a reverse proxy and network monitoring tool; supaquiz, a realtime multiplayer trivia game; Lookout, a metadata-based image discovery app; and Vista, a video transcoder that adds customizable subtitles from audio. Winners received keyboards or swag, and a Discord event was scheduled for January 5 at 7:00 AM PST.
Supabase is releasing Custom Domains and Vanity Subdomains to give projects a more branded user-facing address. Custom Domains let a project use its own domain, such as api.example.com, instead of an assigned supabase.co hostname; they are generally available as a paid add-on through the Supabase dashboard. Vanity Subdomains let paid-plan projects choose a relevant subdomain on Supabase's domain, such as my-example-brand.supabase.co, at no additional cost, and the feature is launching in closed beta with setup instructions in the documentation. The release currently supports only one custom domain or vanity subdomain per project, while Edge Functions must still be accessed through their existing endpoint because they do not support either feature yet.
Oxygen is Shopify’s hosting counterpart for Hydrogen custom storefronts, designed to provide globally available, fast storefronts with secure Shopify integration without additional third-party hosting costs. Its architecture combines GitHub and GitHub Actions with Cloudflare Workers and edge distribution, Shopify CDN, Admin API, Identity and Access Management, and Oxygen-scoped developer tooling and observability. Requests pass through a Gateway Worker, which validates authorization for the shop and storefront version before using Cloudflare’s Dynamic Dispatch API to route traffic to a Storefront Worker running Hydrogen code. Rather than build proprietary infrastructure, Shopify reused Cloudflare’s worker platform to reduce upfront complexity, scale, and security risks, while its observability stack uses Grafana, Cortex, Loki, and Tempo. Oxygen is used in production by Allbirds, Shopify Supply, Shopify Hardware, and Denim Tears, and is generally available to Plus merchants.
Supabase CLI v1 is moving out of beta with capabilities for managing database migrations and generating TypeScript types, alongside a beta Management API. The CLI can manage organizations and projects, link projects locally, run schema and manual migration workflows, support CI/CD through local and GitHub Actions workflows, and generate types from local or remote databases. Migration changes include improved manually written migrations, an experimental migra-backed db diff that is faster and produces more concise DDL, and local branching for testing DDL without data loss; the source notes that generated scripts are not perfect. The REST-based Management API, previously called the Admin API, is consumed by the CLI and exposes programmatic management of organizations, projects, Edge Functions, and more, with endpoints intended for CI/CD and test-environment creation.
This tutorial builds a personal remote development environment: an in-browser IDE for Elixir/Phoenix work, deployed as a Fly Machine from a Dockerfile and managed with command-line tools. The image uses an official Elixir base, adds code-server, flyctl, build dependencies, and the Elixir Language Server extension, while a multi-stage build copies in Tired Proxy. A persistent volume holds working files, and the startup configuration can clone a repository, launch code-server behind HTTPS, and arrange for the machine to sleep after an hour without HTTP requests. From the IDE terminal, the walkthrough then creates and deploys a separate Phoenix hello-world application backed by SQLite, setting its secret and provisioning storage. The resulting setup can wake on browser access, but random HTTP traffic or an active code-server terminal may keep it running, and stopping or destroying the apps limits costs.
Fly.io's June 23, 2022 logbook records a broad set of platform, dashboard, and flyctl updates. It announces YUL (Montreal, QC, Canada), adds Laravel and Laravel Octane detection to fly launch, and improves plain Docker build output when NO_COLOR=1 is set. Backend work optimizes API queries, reduces repeated app-image loads, moves app-discard cleanup into a retryable job, and adds SQL comments to help investigate database data-lock issues. The release also documents diagnostics, makes compliance documents public to prospective customers, and improves fly-proxy’s Rust coverage and PROXYPROTO v2 TLV support. Fixes address fly.toml environment variables, organization deletion, password-reset redirects, app secrets listing, metrics after fly apps move, dashboard errors, malformed SSL TLVs, and overzealous email risk checks.
Fly.io’s June 10, 2022 logbook lists platform, dashboard, deployment, and account-management changes for its globally distributed application hosting service. WireGuard peer synchronization now adds only peers whose state changed, while removing a global lock from proxy tracing instrumentation improves TLS handshaking and HTTP parsing under load; the post says these changes should speed userspace features, avoid API timeouts, and reduce response times. Deployments gain build-time secrets through fly deploy’s --build-secret flag, larger rolling-deploy batches that are expected to make multi-region releases 2–5x faster, and restored Turboku webhook redeployment from Heroku. Dashboard updates include intelligent longpolling fallback, visible certificate deletion controls, monitoring links, safer app deletion confirmation, and corrected metrics and coupon visibility, while the Jupyter launcher now requires a credit card after abuse.
Threaded servers such as Puma and thread-based Sidekiq are popular Ruby deployment choices because threads can increase throughput without a proportional increase in memory. The post compares process and thread models using a memory model that separates static application memory from per-request processing memory, then refines it with Copy on Write (CoW), where forked processes share pages until writes force copies. It explains why RSS can double-count shared pages, recommends PSS and Linux’s /proc/$PID/smaps_rollup for more meaningful measurements, and gives a production example in which about two-thirds of static memory is shared. The conclusion is metric-dependent: preload applications to improve CoW, prefer fewer larger containers when appropriate, consider Unicorn for reliable request timeouts, and start web workloads at two threads unless substantial I/O wait justifies more; job processors may benefit from higher counts.
Fly.io’s April 20, 2022 logbook combines a recent changelog with an account of how the company produces updates for a distributed platform running apps in VMs on hardware around the world. It records added IAD capacity, backup regions disabled by default, flyctl releases with NuxtJS and RedwoodJS deployment support, and reliability improvements to postgres operations, failover handling, error reporting, builders, logs, and releases. The post says commits and status updates often fail to capture completed user-relevant work or its impact. It therefore asks engineers to write updates that explain both what changed and why it matters. Collection is automated through a Slack bot and an app that collates entries into feeds, while the team plans to cultivate stronger changelog habits.
Production deployments of BERT and Transformer architectures often face cost and latency challenges because these models are significantly larger and more computationally intensive than traditional algorithms. To optimize text classification workloads, developers can compile vanilla PyTorch models for AWS Inferentia using the AWS Neuron SDK and its tracing utilities. Because the Neuron SDK requires static tensor dimensions, the model is traced with fixed input lengths and packaged alongside a custom inference script configuring one Neuron Core per worker. Deploying the resulting artifacts to an Amazon SageMaker ml.inf1.xlarge endpoint yields an average latency of 5 to 6 milliseconds for a sequence length of 128 across 10,000 synchronous evaluation requests.
Deploying EleutherAI's 6 billion parameter GPT-J model for production inference presents latency hurdles due to large memory footprints and lengthy startup times. Loading the model via standard methods takes up to several minutes, conflicting with strict real-time response limits such as Amazon SageMaker's 60-second threshold. To overcome this limitation, the model is serialized using PyTorch's native save mechanisms, packaged into a compressed archive with supporting assets, and stored on Amazon S3. This alternative loading workflow reduces GPT-J load times down to 7.7 seconds. An Amazon SageMaker real-time endpoint is then deployed on an NVIDIA T4 GPU instance using the Hugging Face Inference Toolkit.
Recent findings from the 2021 State of AI Report and Kaggle State of Machine Learning and Data Science Survey indicate that machine learning is expanding into critical infrastructure while Transformers become general-purpose architectures across text, vision, and audio. In response, organizations face questions about scaling infrastructure, team composition, and engineering maturity. Rather than treating machine learning as isolated sandbox experiments or hiring solely data scientists, teams benefit from adopting established software engineering and DevOps principles like versioning, testing, automation, and continuous deployment. Furthermore, the rise of pre-trained Transformer architectures enables practitioners to fine-tune existing off-the-shelf models rather than training from scratch, reducing compute costs and training duration. Tools from platforms such as Hugging Face streamline model deployment, latency optimization, and infrastructure abstraction.
YJIT’s benchmarking site combines open-source benchmark runners with Jekyll, GitHub Pages, and GitHub Actions to publish frequently refreshed results. The author wanted to compare Ruby configurations, expand reporting beyond console output, and avoid recurring operational work. Benchmarks run on an AWS instance rather than in GitHub Actions, while Actions schedules jobs, uses repository secrets for SSH, and commits generated data back to the repository; Jekyll builds the Pages site from that data. Reports include D3.js and Ruby-generated SVG graphs, tables, textual summaries, and JSON tripwire data that can trigger an issue when performance drops too far. The reported results show YJIT 20.6% faster than interpreted CRuby and 17.4% faster than MJIT, with Railsbench gains of 18.5% and 21.0% respectively.
Streamlit enables developers to create interactive web demonstrations and visualizations for machine learning models and datasets. In an illustrative text completion interface recreating Write with Transformer, UI components such as text areas, sidebars, number inputs, and continuous sliders collect user text and inference hyperparameters. For data inspection, the streaming feature in Hugging Face Datasets loads large corpora directly into memory without requiring complete local downloads, integrating with pandas, matplotlib, and seaborn for tabular and chart rendering. Deploying these interactive applications to Hugging Face Spaces involves uploading the code files while defining dependencies inside a requirements.txt file. Matching local Streamlit versions ensures consistent runtime execution when hosting models and datasets online.
Demonstrating machine learning projects to the community requires accessible hosting and minimal interface boilerplate. Gradio integrates directly with the Hugging Face Hub and its underlying Inference API, enabling developers to load models using repository identifiers and launch interactive interfaces with a few lines of code. Hugging Face Spaces provides a free hosting environment where users select Gradio as the SDK, commit an app.py script, and deploy live web demonstrations for frameworks like Transformers, spaCy, SpeechBrain, and Asteroid. For complex workflows, developers can chain multiple models sequentially via Gradio Series or host custom checkpoints directly when models lack native Inference API support.
Hugging Face launched its Hardware Partner Program at the 2021 AI Hardware Summit to offer device-optimized models and software integrations. As a founding member of the program, Graphcore collaborated with Hugging Face to optimize Transformer models for its Intelligence Processing Unit (IPU). Graphcore IPUs use a massively parallel MIMD architecture with on-die ultra-high bandwidth memory designed for AI workloads featuring fine-grained parallelism, low precision arithmetic, and sparsity. Through Hugging Face's open-source Optimum toolkit, developers will receive plug-and-play access to certified, hardware-optimized models across vision, speech, translation, and text generation. Graphcore's Poplar SDK integrates with PyTorch, TensorFlow, Docker, and Kubernetes, simplifying the process of porting workloads to IPU-POD datacenter compute systems.
Developer Infrastructure at Shopify spans local and cloud development environments, CI systems, deploys, and technology-specific foundations for web and mobile applications. The post models these teams against a three-phase develop–validate–deploy workflow, distinguishing horizontally integrated teams that own one phase from vertically integrated teams whose systems span phases for a particular technology. It uses this model to identify dependencies, gaps, differing scopes, collaboration patterns, and three work avenues—concentration, expansion, and interfacing—with the mission of improving and scaling the cycle. The analysis informed organizational changes, including a Client Foundations team, closer collaboration, and plans for workflow-wide user research focused on the developer journey from writing a patch to production.
The best engineering reads, once a week.
