BEGIN:VCALENDAR VERSION:2.0 PRODID:-//HasGeek//NONSGML Funnel//EN DESCRIPTION:May 2024 Rustacean meetup X-WR-CALDESC:May 2024 Rustacean meetup NAME:May 2024 Rustacean meetup X-WR-CALNAME:May 2024 Rustacean meetup REFRESH-INTERVAL;VALUE=DURATION:PT12H SUMMARY:May 2024 Rustacean meetup TIMEZONE-ID:Asia/Kolkata X-PUBLISHED-TTL:PT12H X-WR-TIMEZONE:Asia/Kolkata BEGIN:VEVENT SUMMARY:Benchmarking Rust Concurrency Libraries DTSTART:20240511T043000Z DTEND:20240511T051000Z DTSTAMP:20260420T095833Z UID:session/9kezP63bZZ5M3QRhb436Qs@hasgeek.com SEQUENCE:8 CREATED:20240503T074403Z DESCRIPTION:An overview of the popular concurrency libraries in Rust\, whi ch were used to generate the Julia fractal. I started out by rendering the fractal using the tokio library so i callled this presentation "Julia in Tokio". I benchmarked them by using them to run a lot of iterations of mul tiple image sizes and used Bemcher to measure average iteration time for t he different implementations. In addition to this\, I generated flame grap hs for each implementation and would be talking about how to read them and make sense of them. The libraries I compared are crossbeam\, tokio\, smol \, rayon. LAST-MODIFIED:20240503T094702Z LOCATION:Fortanix\, HSR Layout\, Bangalore ORGANIZER;CN="Rust Bangalore":MAILTO:no-reply@hasgeek.com URL:https://hasgeek.com/rustbangalore/may-2024-rustacean-meetup/schedule/b enchmarking-rust-concurrency-libraries-9kezP63bZZ5M3QRhb436Qs BEGIN:VALARM ACTION:display DESCRIPTION:Benchmarking Rust Concurrency Libraries in 5 minutes TRIGGER:-PT5M END:VALARM END:VEVENT BEGIN:VEVENT SUMMARY:Introducing Superposition - a flexible\, safe and powerful Open So urce Configuration Management System DTSTART:20240511T051000Z DTEND:20240511T055000Z DTSTAMP:20260420T095833Z UID:session/Htd3S7fi7gXNyCxWVVWTP6@hasgeek.com SEQUENCE:6 CREATED:20240503T074419Z DESCRIPTION:Most non-trivial applications need a robust and flexible confi guration system. Such a system becomes all the more important when the sam e feature needs to behave differently for different users. At Juspay\, we needed a configuration system with the following properties: \n\n - First\ , it should be flexible\, allowing teams to make changes quickly without a pplication deployment. \n- Second\, it should be Safe - ensuring that the value provided for a configuration is type correct\, and the rollout of su ch a change is staggered. \n\nWe looked around for off-the-shelf software and did not find anything that met these requirements\, so we built Superp osition.\n\nSuperposition is a context-based configuration management syst em designed to facilitate safe and flexible rollout of config changes. \n\ nThis is enabled by two key components of Superposition.\n- Context Aware Config (CAC) - Lets you do granular configuration based on contextual data .\n- Experimentation - Lets you do A/B configuration deployment\, allowing controlled rollout and rollback mechanisms.\n\nFrom discussions with mult iple teams\, we feel our requirements for a configuration system are unive rsal. Teams inside Juspay like Juspay Portal\, Hyper SDK and Namma Yatri h ave started to leverage Superposition to manage their complex configuratio n use-cases in production.\n\nWe have open sourced Superposition recently\ nhttps://github.com/juspay/superposition\n\n## Key highlights\n* **Admin U I** - Out of the box administration (and tools) UI for configurations and experiments\n* **Rich API support** - every action on the platform to mana ge configurations / experiments is supported with an accompanying API\n* * *Safe configuration changes** - support canary testing for releasing confi guration changes using experiments\n* **Type/Validation support** - Compre hensive type support using json-schema and custom validator function suppo rt for configuration values\n* **Multi-tenant support** - a single deploym ent allows multiple tenants to manage their configurations/experiments in a completely isolated manner\n* **Authn/Authz support** - control who can make configuration/experimentation changes LAST-MODIFIED:20240503T074459Z LOCATION:Fortanix\, HSR Layout\, Bangalore ORGANIZER;CN="Rust Bangalore":MAILTO:no-reply@hasgeek.com URL:https://hasgeek.com/rustbangalore/may-2024-rustacean-meetup/schedule/i ntroducing-superposition-a-flexible-safe-and-powerful-open-source-configur ation-management-system-Htd3S7fi7gXNyCxWVVWTP6 BEGIN:VALARM ACTION:display DESCRIPTION:Introducing Superposition - a flexible\, safe and powerful Ope n Source Configuration Management System in 5 minutes TRIGGER:-PT5M END:VALARM END:VEVENT BEGIN:VEVENT SUMMARY:Fortanix Enclave Development Platform (EDP)\, an open-source platf orm for developing Confidential Computing Applications DTSTART:20240511T061500Z DTEND:20240511T065500Z DTSTAMP:20260420T095833Z UID:session/9qi169Jve9k9vxSoZAskrQ@hasgeek.com SEQUENCE:2 CREATED:20240503T074515Z DESCRIPTION:Fortanix Enclave Development Platform (EDP)\n\nProblem Space: \nData exists in three states: at rest\, in use\, and in transit. Data tha t is stored is "at rest"\, data that is being processed is "in use"\, and data that is traversing across the network is "in transit". Generally\, or ganizations focus on protecting data at rest and in transit across the net work. However\, the data being processed is still vulnerable to unauthoriz ed access and tampering at runtime. Protecting the data in use is critical to offer complete security across the data lifecycle.\n\nSolution:\nConfi dential Computing is an approach that uses secure enclave technology to en able the creation of a Trusted Execution Environment (TEE) based on securi ty features provided by CPU vendors.\nThis provides protection of data bei ng processed through using hardware-based TEEs.\nConfidential computing ap plications achieve this by running in such secure enclaves.\n\nGenerally t here are two ways to develop Confidential Computing applications -\n1. Tak e an existing application and convert it to be enclave compatible.\n2. Dev elop applications in a platform that inherently supports enclave compatibi lity.\n\nHow Fortanix provides Confidential Computing -\nApproach-1 mentio ned above\, is non-trivial and comes with disadvantages such as not being able to access underlying FS or network which the existing application may well be depending upon.\nFortanix Enclave Development Platform (EDP) take s this 2nd approach \nand is the preferred way to write Intel SGX enclaves from scratch in Rust language.\nThe platform provides some utility tools as well such as 'sgx-detect' \nthat detects if the SGX technology is avail able on a given system\, \nand if so\, detects and provides information on various components relating to SGX.\n\nWe'll showcase building an applica tion to the EDP specific target and discuss the restrictions it comes with .\n\nopensource link: https://github.com/fortanix/rust-sgx LAST-MODIFIED:20240503T074556Z LOCATION:Fortanix\, HSR Layout\, Bangalore ORGANIZER;CN="Rust Bangalore":MAILTO:no-reply@hasgeek.com URL:https://hasgeek.com/rustbangalore/may-2024-rustacean-meetup/schedule/f ortanix-enclave-development-platform-edp-an-open-source-platform-for-devel oping-confidential-computing-applications-9qi169Jve9k9vxSoZAskrQ BEGIN:VALARM ACTION:display DESCRIPTION:Fortanix Enclave Development Platform (EDP)\, an open-source p latform for developing Confidential Computing Applications in 5 minutes TRIGGER:-PT5M END:VALARM END:VEVENT BEGIN:VEVENT SUMMARY:Monitoring sytems with Phi Accrual Detectors in Rust DTSTART:20240511T065500Z DTEND:20240511T073000Z DTSTAMP:20260420T095833Z UID:session/KmPwCPZ9qPfUbZZmegQYgb@hasgeek.com SEQUENCE:4 CREATED:20240503T074535Z DESCRIPTION:# whoami\nI am Uddeshya Singh ([X](https://twitter.com/uds5501 )\, [Medium](https://medium.com/@singhuddeshyaofficial)\, [LinkedIn](https ://www.linkedin.com/in/singhuddeshya5501/))\, SDE-2 Backend at Gojek. I sp ecialize in payments in my 9-5 and have been an avid distributed systems + database geek. \n\n# Abstract\nIn distributed systems\, ensuring reliabil ity and fault tolerance is of utmost concern. Traditional heartbeat failur e detectors often fall short in accurately assessing the health of process es\, especially in environments with transient failures or network partiti ons. This is where Phi Accrual Failure Detectors (φ-failure detectors) co me into play\, offering a more nuanced and adaptive approach to failure de tection.\n\nRust is an amazing choice for distributed systems so I decided to try and build a crate for the same\, which is pluggable directly into your raft / paxos / toy redis systems. **The result?** [crates.io/phi-accr ual-detector](https://crates.io/crates/phi-accrual-detector).\n\n# Top tak eaways from the talk\nWhat would the audience learn after this talk?\n- Bu ilding an intuition regarding why heartbeats might not be the best usecase always for failure detection.\n- What makes a failure detector amazing?\n - Comparing various detection algorithms and finally elaborating on φ-fai lure detectors.\n- Code walkthrough for implementation of the library and test run results.\n LAST-MODIFIED:20240503T074611Z LOCATION:Fortanix\, HSR Layout\, Bangalore ORGANIZER;CN="Rust Bangalore":MAILTO:no-reply@hasgeek.com URL:https://hasgeek.com/rustbangalore/may-2024-rustacean-meetup/schedule/m onitoring-sytems-with-phi-accrual-detectors-in-rust-KmPwCPZ9qPfUbZZmegQYgb BEGIN:VALARM ACTION:display DESCRIPTION:Monitoring sytems with Phi Accrual Detectors in Rust in 5 minu tes TRIGGER:-PT5M END:VALARM END:VEVENT END:VCALENDAR