Datadog Blog

And it's worth noting that dalmatiner magic relies on ZFS + riak. The ZFS part being problematic to run on linux. There is currently no open source monitoring solutions that support more than a single node. If you have serious infrastructure, you gotta move to the paid SaaS monitoring solutions which are ALL awesome.: https://www.datadoghq.com/ and https://signalfx.com/ Note: I am not affiliated with any of these services. I just have a very strong interest in nagios/icinga/riemann/graphite/grafana dying since my painful experience of setting up and maintaining them.

If you can have a monitoring system in the cloud Datadog is a great choice. Good documentation, UI, many, many plugins and fair pricing (IMO). https://www.datadoghq.com/ (Im not affiliated with in any way other than using their product on a pet project with many moving parts).

Datadog - http://www.datadoghq.com/ - New York, NY Generalists, front-end, back-end & data developers. Datadog is a product built to give dev and ops teams great visibility into the health of their system. If you want to come and build systems that process persist, analyze and visualize tons of data (millions of data points per second), come and check us out. For me, a backend dev, it's been incredibly fruitful helping build, grow and run a system at this scale. We've tackled lots of hard problems and there are lots left to solve. More details here: http://jobs.datadoghq.com/ Drop me a line if you have any questions: [email protected]

Personally and professionally, I have had a great experience with Datadog (https://www.datadoghq.com/) - plenty of integrations with various services, providers and a useable interface. Prior to DD, I used ServerDensity (https://www.serverdensity.com/) - also a very mature tool and the technical blog written by their founder (https://blog.serverdensity.com/) has interesting tidbits from time to time. I'm sure someone else will mention New Relic, too - great tool, but more expensive than the other two (though well worth it, if you need it)

Datadog (http://www.datadoghq.com) New York, NY - Full-time python hackers (pandas, scipy, numpy, hadoop), devops engineers (aws, haproxy, cassandra, kafka, redis, etc.) and support engineers. tl;dr We are a monitoring service, processing hundreds of billions of records every day. Hard problems, great team, paying customers, real-time python stack. What makes us different? Datadog is a product for you, developers, webops. So when you join Datadog, you have a strong and direct voice in the making of the product. Interested? Apply at http://jobs.datadoghq.com

Hey Nick - come and visit us at Datadog http://datadoghq.com We're on 26th and Broadway, next to a great many other startups. I'm oli@

Datadog | Software Engineers | ONSITE (Boston, Lisbon, Madrid, NYC, Paris, Tel Aviv) | Full-time Datadog is a monitoring, tracing, logs system, and more, for your infrastructure and services. We build our own tsdb, event store [1][2], distributed tracing tools, cutting edge visualizations, and more. We love shipping great experiences for customers just like us and are growing fast! We write a lot of Go, Java, Python, Typescript (with React), and a bit of other languages. We run on k8s, and are multi-region and multi-cloud. We're looking for people who can build systems at scale as we process trillions of events per day. Let us know if that's you! https://dtdg.co/hnwhoshiring [1] https://www.datadoghq.com/blog/engineering/introducing-husky [2] https://www.datadoghq.com/blog/engineering/husky-deep-dive

> Link me the research on the hard engineering tasks they've done on database kernels, I'd love to see it, sounds interesting. https://www.datadoghq.com/blog/ai/harness-first-agents https://www.datadoghq.com/blog/ai/fully-autonomous-optimization https://www.datadoghq.com/blog/engineering/self-optimizing-system

This post on Glommio is insightful - "Glommio is a cooperative thread-per-core crate for Rust & Linux based on io_uring": https://www.datadoghq.com/blog/engineering/introducing-glommio/ The key insight being that in this model "locks are never necessary", which is huge source of efficiency when working with sharded data across multiple cores. > Thread-per-core architectures are friendly to modern hardware, as their local nature helps the application to take advantage of the fact that processors ship with more and more cores while storage gets faster, with modern NVMe devices having response times in the ballpark of an operating system context switch.

Datadog | Senior Software Engineer | NYC/Paris/Madrid | Full-Time | Static Analysis We are building a static analysis engine (https://github.com/DataDog/datadog-static-analyzer) that works across the development lifecycle: IDE -> git hook -> git pull request -> git push. Our product is currently in private beta (see https://www.datadoghq.com/static-analysis/) and we are working hard on expanding its capabilities. If you work with Rust and/or Tree-Sitter, we would love to talk to you! Apply here > https://careers.datadoghq.com/detail/5382867/?gh_jid=5382867

> and likewise with any compute task that can be scaled out instead of up. So move as much as possible out of the scale-up instance to scale-out storage and compute. This was very much the main lesson I got reading up on DataDog's third-gen event storage system Husky[1]. Great read. [1] https://www.datadoghq.com/blog/engineering/introducing-husky/ https://news.ycombinator.com/item?id=31416843 (185 points, 9d ago, 15 comments)

You don’t need to go this path; modern(-ish) APM systems like Datadog will let you trace SQL commands in the context of the request being made. You can group by endpoint, or group by sql (merging IDs that differ), or chop and filter the data however you need. The DB is not in a position to give you all the stats you need; instead the DB should support e.g. exporting plan info to further enrich the whole-request metrics that are already being collected. I use this in prod; it’s great. Expensive though. https://www.datadoghq.com/blog/mysql-monitoring-with-datadog/

Can you add a little more color commentary? I haven't used datadog, but they seem clear that there's both a logging and metrics (apm) side to their product. Is the logging side neglected somehow? I see, for example, docs on how to log from Python: https://www.datadoghq.com/blog/python-logging-best-practices/

>> While the founders, CEO Martin Mao and CTO Rob Skillington, were working at Uber, they recognized a gap in the monitoring industry, particularly around cloud-native technologies like containers and microservices What is the actual gap that is not addressed with one or all of the following? - Prometheus / Grafana [1] - Datadog [2] - Cloudwatch [3] 1. https://docs.docker.com/config/thirdparty/prometheus/ 2. https://www.datadoghq.com/blog/introducing-live-container-monitoring/ 3. https://docs.aws.amazon.com/AmazonCloudWatch/latest/monitoring/Contain…

> Optional: verify public key on Keybase. For organizations publishing employee keys via Web Key Directory can also be an additional signal that the key is trustworthy. It's also quite simple: for example exporting the key 5C090ED7318B6C1E (binary, not armored) and putting it on this exact URL: https://datadoghq.com/.well-known/openpgpkey/hu/964aj6q73iatngoya1q7qs… is enough to discover the key using e-mail address. This post goes into more detail: https://spacekookie.de/blog/usable-gpg-with-wkd/ WKD is used by Linux distros (ArchLinux, Gentoo, Debian...) and kernel.org itself: https://www.kernel.org/category/signatures.html#using-the-web-key-dire… as well as some OpenPGP sites (e.g. ProtonMail).

I am very keen on picking up some $DDOG shares tomorrow when it IPO'S. As part of my DevOps consultancy I recommend DataDog to nearly all my clients. They check all the boxes in one unified platform: [X] Server / cloud resource monitoring and alerting [https://www.datadoghq.com/dashboarding/] [X] Logging [https://docs.datadoghq.com/logs/] [X] APM (application performance) [https://docs.datadoghq.com/tracing/] [X] HTTP/API endpoint checks and alerts [https://docs.datadoghq.com/synthetics/]

I work at Datadog and we've been pretty successful at attracting customers over from Splunk depending upon the use case. Others in industry are SumoLogic, ELK etc. You can check us out here if you want https://www.datadoghq.com/

Datadog | Software Engineers | ONSITE (Boston, New York City, Paris) and REMOTE | Full-time Datadog is a monitoring and tracing service for your infrastructure and services. We build our own tsdb, distributed tracing tools, cutting edge visualizations, and more. We love shipping great experiences for customers just like us and are growing fast! We're mostly Go, Python, and React, on AWS, and moving to k8s. We're looking for people who can build systems at scale as we process trillions of events per day. Let us know if that's you! https://www.datadoghq.com/jobs-engineering/?gh_src=ywmco2h41

Terraform + Datadog + Cloudwatch http://terraform.io/ http://datadoghq.com/ https://aws.amazon.com/cloudwatch/

DataDog[1] + PagerDuty[2] do a good job at being linux-doctor. You have to be pro-active about it though... [1] http://www.datadoghq.com/ [2] http://www.datadoghq.com/

We use DataDog - http://www.datadoghq.com/ . It’s a statsd, graphite manifestation but with much more capabilities. Stack is Python, Django, PostgreSql, Redis, Memcache etc.

> If that were true then they wouldn't be heap allocations. "Heap" is a misnomer. It's not called that due to the classic CS "heap" datastructure. It's called that for the same reason it's called a heap allocation in C++. Modern C++ allocators don't use a heap structure either. How the JVM does allocations for all it's collectors is in fact a bump allocator in the heap space. There are some weedsy details (for example, threads in the JVM have their own heap space for doing allocation to avoid contention in allocation) but suffice it to say it ultimately translates into a region check then pointer bump. This is why the JVM is so fast at allocation, much faster than C++ can be. [1] [2] > I don't know what this means. JVM allocations are typically pointer bumps, adding a number to a register. There's really nothing faster than it. If you are implementing an arena then you've already lost in terms of performance. [1] https://www.datadoghq.com/blog/understanding-java-gc/#memory-compactio… [2] https://inside.java/2020/06/25/compact-forwarding/

Very cool, and very much what seems to be the trend for very high performance data systems: disaggregated storage. I think of Datadog's write-up for Husky, their newest/3rd gen data engine; https://www.datadoghq.com/blog/engineering/introducing-husky/ > By writing data directly to S3-compatible object storage, Bufstream can dispense with expensive disks and nearly eliminate inter-zone networking, slashing costs dramatically. I am curious what the situation looks like for self-cloud-ers. If you own the servers, its less clear how much advantage there would be using s3 object store versus disk attached storage services. But reciprocally, getting good at Ceph and Ceph Object Gateway (their s3 compat) then being able to use and tune that storage knowledge at the platform level makes sense, versus having a separate storage service for x, y, and z. Still, I think there is huge potential for something like Pulsar with BookKeeper tablets of data to rise. Our object stores don't seem to be excellent at data-locality, at replicating, and being able to tap that could yield some incredible systems efficiencies & speeds that object storage has abstracted away from us, that object storage has to brute force. I'm curious what architecture flourishes there are with your latency tuning. Is this still object store based, or something different? > For particularly latency-sensitive or latency-tolerant workloads, operators can tune how aggressively Bufstream trades latency for cost. The…

https://www.datadoghq.com/ has something like that

Datadog | Full-time, Permanent Roles: Data Engineer | Software Engineer - Data Infra | Distributed Systems Engineer (ALL AVAILABLE IN REMOTE US/NYC/BOSTON) 1) Data Engineer: Looking for engineers to work on realtime (Kafka + Kafka Connect + custom Go consumers) and batch systems (Spark on EMR + Dataproc + Luigi + Parquet + S3 and GCS) to process 100's of TBs daily - at times PB-scale per day. Apply here for BOSTON/NYC - https://grnh.se/f4127d5f1 - if REMOTE US, email [email protected]. 2) Software Engineer - Data Infra: Engineers in our Data Infra team build and own all infrastructure required to transport, process and store data at scale. This team also owns the system that allows teams to schedule and execute their batch jobs on multiple cloud platforms in multiple regions. Apply here for BOSTON/NYC/REMOTE US - https://grnh.se/ef2ed6f51. 3) Distributed Systems Engineer: Help us build the high-throughput, low-latency systems that power our product. These systems ingest, store, analyze, and query tens of millions of events per second from companies all over the globe. Looking for experience in Go and Python, with bits of C or other languages. Apply here for BOSTON/NYC - https://grnh.se/866fe22c1 - if REMOTE US, email [email protected]. More details about Datadog here: https://www.datadoghq.com/

Thank you very much for your comment - I've realized that caching DNS should be considered in any "connected" service. I would add my two cents: - Calculate checksum and compare it with the latest entry (storing single 32/40-bytes long checksum takes lower amount of memory). - Respect various HTTP headers like Expires, Cache-control, Accept-Encoding (accepting compressed content may relieve strain on bandwidth), Content-length, Cookie-length etc. - If you were downloading a web-page and sending it via e-mail (for later usage?) you may gzip it - it may reduce costs of attachments or S3. - Attach header "User-Agent" and "From" (with an e-mail address to contact developer) - if you don't mind. - As you wrote, avoiding ioutil.ReadAll should be a good practice - Datadog published a blogpost which is worth reading - https://www.datadoghq.com/blog/crossing-streams-love-letter-gos-io-rea… - Use logrotate or implement it on your own with i.e. Mutex. Eventually you use fluentd, Kafka, Pub/Sub pipeline or some SaaS offer specialized in storing logs. Most (or all) of Go packages with "request" in the name that I know doesn't cover aforementioned scenarios automatically - it is up to developer.

Ilan with Datadog here. On Friday, September 30th a number of Datadog-AWS shared customers received an erroneous email notification from AWS about compromised AWS access keys associated with Datadog. AWS intended to reach out to shared customers that had been lax in deactivating and deleting access keys associated with the 7/8/2016 Datadog breach (https://www.datadoghq.com/blog/2016-07-08-security-notice/). AWS accidentally sent out a standard compromise notification to the original list of shared customers. False positives abound! You can validate this error by contacting your AWS support contact. As recommended in our original communication, if you have not you should deactivate and delete any service integration credential shared with Datadog on or prior to 7/8/2016 immediately and for AWS transition to Role Delegation as outlined here: https://help.datadoghq.com/hc/en-us/articles/210376966-Revoking-AWS-ke… The AWS communication was made in error. If you received the initial message you should have also received a retraction. Customers that have been lax in deactivating and deleting these access keys will receive additional communication directly from AWS.

I'm a technical writer at Fastly and a published author with No Starch Press. I've worked full time as a technical writer since 2006. It's a great career and I highly recommend it. If you have a technical background and you've coauthored a book, you already have what it takes to find full time work as a technical writer. There are really two ways into this field. You start with a background in English or professional writing and learn the technology, or you can start with a computer science degree or something similar and learn how to write. It sounds like you've done the latter. The pay is good if you work for a company full time. Many technical writing jobs these days are remote. (All of the tech writers that work for Fastly and GitHub are remote, for example.) You could probably do freelance work, but I've never been able to make that work. Most companies don't outsource their documentation and I'm not entirely sure why, although I suspect it has something to do with them wanting to build long term relationships with the individual(s) documenting their products. You will need to do some networking. You may want to consider attending Write the Docs (http://www.writethedocs.org/), a conference for technical writers. Depending on where you live, you might also be able to find an active STC group. Or just start applying for jobs. :) Really good technical writers are hard to come by, so if you know technology and you can write, you'll probably have the job offers flooding…

I agree, New Relic's pricing model is almost too "enterprise". Per host pricing seems high, and difficult to scale. It's too bad, because I love everything else they offer. I'm even considering the new Insights service. I think they need to introduce a mid-tier pricing level between Free and Pro. They've got a good foothold in the market as a all-in-one solution. I've attempted to find a drop-in replacement, without much luck. It's difficult to find a one-for-one service, but here's some similar services with entry-level pricing: DataDog (https://www.datadoghq.com/pricing/): $15/host/month up to 100 hosts. Scout (https://scoutapp.com/subscriptions): $10/host/month CopperEgg (http://copperegg.com/pricing/): A la carte depending on service Stackify (http://www.stackify.com/pricing/): $15/host/month The biggest feature that I need is application level monitoring for PHP. There's plenty of great server monitoring tools, but not many services that track application performance like New Relic does. I know there's plenty of ways to replicate New Relic, but who has the time? Of course, I could be completely wrong, and missing a great alternative service.

Have you looked at http://www.datadoghq.com/