napkin-math
share-file-systems
napkin-math | share-file-systems | |
---|---|---|
13 | 34 | |
3,093 | 125 | |
- | - | |
6.3 | 8.7 | |
12 days ago | 3 months ago | |
Rust | TypeScript | |
MIT License | GNU Affero General Public License v3.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
napkin-math
- capacity planning in system design interviews
- Napkin Math
-
S3 Express Is All You Need
Most production storage systems/databases built on top of S3 spend a significant amount of effort building an SSD/memory caching tier to make them performant enough for production (e.g. on top of RocksDB). But it's not easy to keep it in sync with blob...
Even with the cache, the cold query latency lower-bound to S3 is subject to ~50ms roundtrips [0]. To build a performant system, you have to tightly control roundtrips. S3 Express changes that equation dramatically, as S3 Express approaches HDD random read speeds (single-digit ms), so we can build production systems that don't need an SSD cache—just the zero-copy, deserialized in-memory cache.
Many systems will probably continue to have an SSD cache (~100 us random reads), but now MVPs can be built without it, and cold query latency goes down dramatically. That's a big deal
We're currently building a vector database on top of object storage, so this is extremely timely for us... I hope GCS ships this ASAP. [1]
[0]: https://github.com/sirupsen/napkin-math
-
Random Read or Sequential Read
Trying to estimate performance using some napkin math based on this: https://github.com/sirupsen/napkin-math
-
A CVE has been issued for hyper. Denial of Service possible
So napkin maths time. Typical cross-world bog-standard network speeds for a single TCP channel of ~25MiBps. A single HEADERS+RST pair is likely < 128 bytes (40 for the HEADERS + whatever payload, and 32 for the RST). So 8 pairs per K, 8K pairs per MiB, 200K pairs per 25MiB...
- Index Merges vs Composite Indexes in Postgres and MySQL
-
I/O is no longer the bottleneck
Yes, sequential I/O bandwidth is closing the gap to memory. [1] The I/O pattern to watch out for, and the biggest reason why e.g. databases do careful caching to memory, is that _random_ I/O is still dreadfully slow. I/O bandwidth is brilliant, but latency is still disappointing compared to memory.
[1]: https://github.com/sirupsen/napkin-math
- Monthly cost to host server for 1M DAUs?
- Napkin-math: Techniques and numbers for estimating system's performance
-
System Design prep?
https://github.com/sirupsen/napkin-math (memorize these)
share-file-systems
-
Lcl.host: fast, easy HTTPS in your local dev environment
Some things I learned about trusted localhost HTTPS:
* Windows is the easiest... by far. There is only one trust store and its extremely easy to access at different levels of trust. Firefox has its own trust store so you can either add your certs to both the Windows store AND the Firefox trust store or flip a config in Firefox to tell it to use the Windows trust store like everyone else.
* Linux is a challenge because you have to add your certificates to the OS trust store and then each browser has their own trust stores.
* MacOS is pretty close to impossible, at least fully automated. If the cert is not registered with a third party of the OS's choosing the cert will not be trusted in the browser. The way around this is to manually add your localhost cert chain to the MacOS keychain.
If anybody wants an example here is something I wrote a ways back in JS (but please be warned its specific to my application:
* Build the certificate chain - https://github.com/prettydiff/share-file-systems/blob/master...
* Install the cert by OS type - https://github.com/prettydiff/share-file-systems/blob/master...
That second sample also installs pcap so that I can serve on localhost over ports 80/443.
-
We have used too many levels of abstractions and now the future looks bleak
Some developers believe everything is always a framework or any attempt to avoid frameworks creates a new framework. I cannot help these people. Any non-religion is a cult type nonsense of affirming the consequent fallacy.
Otherwise a valid example is this one file that creates a complete OS-like GUI in the browser awaiting content typically populated from WebSocket messaging: https://github.com/prettydiff/share-file-systems/blob/master...
-
Os.js – open-source JavaScript web desktop platform with a window manager
I wrote a similar concept around private internet access to your file system. It’s at https://github.com/prettydiff/share-file-systems
The window and state management can be demoed on my personal site at https://prettydiff.com
-
Ask HN: Tell us about your project that's not done yet but you want feedback on
File sharing and soon remote execution over the internet cross OS. Private and no servers.
https://github.com/prettydiff/share-file-systems
-
Meta Forced to Reveal Anonymous Facebook User's Identity
Done: https://github.com/prettydiff/share-file-systems/blob/master...
You would need a warrant to extract the messages/identity directly from a person's computer as there is nothing otherwise to obtain.
-
More encryption means less privacy (2016)
Perhaps this is true in the context of the web. But I got tired of watching the web as a platform continuously repeat the same mistakes so I started working on something different. In the last day or two I was finally able to functionally prove my competing idea in a way that forcefully imposes privacy with complete Zero Trust conformance.
https://github.com/prettydiff/share-file-systems/blob/master...
-
Bfs 3.0: The Fastest Find Yet
I am performing a similar file system tree navigation asynchronously in Node.js which is just a shallow API over the C Linux FS APIs.
I can see you are using opendir and closedir functions? What is the benefit from using the opendir function[1] when readdir[2] can be called on a location directly? Is the benefit that opendir returns a file descriptor for use in opening a stream to gather directory object descriptors?
[1] https://man7.org/linux/man-pages/man3/opendir.3.html
[2] https://man7.org/linux/man-pages/man3/readdir.3.html
Your project is probably more mature but if you want an alternate approach to examine here is I have been doing it: https://github.com/prettydiff/share-file-systems/blob/master...
I considering changing my use of readdir to use the withFileTypes option so that it returns a list of directory entries (objects of artifact name and type) instead of a list of conditions to discern types like I am doing on lines 382-432.
-
Easy HTTPS for your private networks
Solved.
Solved for both Windows and Linux (Debian, Arch, Fedora). I might have unlikely solved this of OSX as well, but I am not buying Apply hardware just to test it.
What my solution does is check for certificates created by the project during a build step. If the certificates don't exist it creates them, installs them in the OS, and also install them in the browser. Installation in the browsers is required in Linux and only for FireFox in Windows. These are cert chains containing a self-signed root, intermediary CA, and a local domain cert.
I have these certs configured to work with my own domains so that I can connect to a subdomain addressed to a loopback IP and the cert recognizes that domain, but the domain "localhost" works as well. Sometimes its nice to access a real domain to avoid any restrictions imposed upon accessing address "localhost". You just have to change the domains at the bottom of your OpenSSL option files.
Here is how I solved it with vanilla TypeScript in Node.js (also requires locally installed OpenSSL:
* OpenSSL option file 1 - https://github.com/prettydiff/share-file-systems/blob/master...
* OpenSSL option file 2 - https://github.com/prettydiff/share-file-systems/blob/master...
* Certificate library - https://github.com/prettydiff/share-file-systems/blob/master...
* Certificate interface from build tool - https://github.com/prettydiff/share-file-systems/blob/master...
* Certificate installation - https://github.com/prettydiff/share-file-systems/blob/master...
If you have any questions just open a Github issue on the project.
-
Ask HN: Who wants to be hired? (June 2023)
Email: [email protected]
15 years experience with JavaScript, 6 years experience with TypeScript. I am currently writing a Node based OS in TypeScript to solve for decentralization (not Web3): https://github.com/prettydiff/share-file-systems
I understand performance aggressively enough far beyond the comfort of most developers: https://github.com/prettydiff/wisdom/blob/master/performance...
-
Ask HN: Are you working on a big software project? Happy with the architecture?
I started a JS based file sharing application a few years back. It started as a thought experiment of just exposing the file system to the browser in a familiar OS kind of user interface. As new features are added over time it has become more like a high level OS.
https://github.com/prettydiff/share-file-systems
Some architectural decisions I made:
* Micro-service based
* I am now using WebSockets for all services and communication. That has proven in the application to be 7x faster than HTTP.
* I have a universal format wrapping all service messaging, kind of like sending a letter in an envelope. This allows me to using a single service end point for all services and a single means of service monitoring.
* I did not like the existing test automation solutions based upon CDP, because they are too slow and fragile. Also, they do not provide support for a peer-to-peer experience. So I wrote my own test automation solution for testing in the browser and its much faster and predictable.
* I am using an identity based authentication mechanism to restrict access to known users/devices.
* I just write to the file system instead of using a database for data storage. This allows for much faster application start up times and lowers complexity. The performance difference is insignificant after accounting for that in most cases opening a file is more costly than arbitrarily writing to the file system.
* I figured out how to install certificates using automation in both Windows and Linux which allows me to run the application using encrypted transmission protocols (https/wss) on localhost.
What are some alternatives?
huniq - Filter out duplicates on the command line. Replacement for `sort | uniq` optimized for speed (10x faster) when sorting is not needed.
DsHidMini - Virtual HID Mini-user-mode-driver for Sony DualShock 3 Controllers
advisory-database - Security vulnerability database inclusive of CVEs and GitHub originated security advisories from the world of open source software.
Redis - Redis is an in-memory database that persists on disk. The data model is key-value, but many different kind of values are supported: Strings, Lists, Sets, Sorted Sets, Hashes, Streams, HyperLogLogs, Bitmaps.
adix - An Adaptive Index Library for Nim
Clendar - Clendar - Minimal Calendar app. Written in SwiftUI.
h2 - HTTP 2.0 client & server implementation for Rust.
userbase - Create secure and private web apps using only static JavaScript, HTML, and CSS.
RAMCloud - **No Longer Maintained** Official RAMCloud repo
circles-ios - E2E encrypted social networking built on Matrix. Safe, private sharing for your friends, family, and community.
simdjson - Parsing gigabytes of JSON per second : used by Facebook/Meta Velox, the Node.js runtime, ClickHouse, WatermelonDB, Apache Doris, Milvus, StarRocks
PhotoPrism - AI-Powered Photos App for the Decentralized Web 🌈💎✨