29 engineers on Hacker News describe using Apache Hadoop first-hand: 5 endorse it, 2 regret it, 3 are mixed.
“They already had Hadoop solutions that more or less worked”
mixed · production · lisbbb · 2025-10-29 · source ↗ · wrong label?“One of the biggest complaints from the business when we switched from SQL server to Hadoop is the terrible tooling.”
regret · production · swalsh · 2018-10-21 · source ↗ · wrong label?“small failures could bring down your entire cluster... I'm not surprised people eventually migrated to distributed systems that were easier to maintain”
regret · production · 2 yr · samspenc · 2017-03-25 · source ↗ · wrong label?“At least with Hadoop you have a little more control over resources with the right configuration.”
endorse · production · SOLAR_FIELDS · 2016-12-19 · source ↗ · wrong label?“So we did it with Hadoop on two Solaris boxes (18 virtual cores between them). Same data; 45 minutes.”
endorse · production · virmundi · 2016-09-12 · source ↗ · wrong label?“The first paying job we ran through our Hadoop cluster in 2011 ... I think that Hadoop wins”
endorse · production · 5 yr · sgt101 · 2016-09-11 · source ↗ · wrong label?“Hadoop is the easiest way for us to do complex processing against our fairly large data set”
endorse · production · angrybits · 2015-07-29 · source ↗ · wrong label?“I like the stack as a whole, I do see the downside of having evolved the stack over time with a bunch of independent projects”
mixed · production · TallGuyShort · 2015-06-10 · source ↗ · wrong label?“it's close to C or C++ for performance and has the advantage of the Hadoop ecosystem behind it”
endorse · production · Smerity · 2014-12-10 · source ↗ · wrong label?“I've seen anywhere from no slowdown to 1000x slower. It sounds to me like those projects should never have been using hadoop”
mixed · production · jeffffff · 2011-11-13 · source ↗ · wrong label?“I was working on different kinds of workloads and more focused on learning Hadoop”
neutral · personal · dnilasor · 2021-09-29 · source ↗ · wrong label?“it is distributed, fault tolerant, and only breaks when the hadoop cluster backs up.”
neutral · production · lykr0n · 2018-08-21 · source ↗ · wrong label?“We use Apache Oozie (http://oozie.apache.org/) an orchestration system for Hadoop.”
neutral · production · jodison · 2018-05-15 · source ↗ · wrong label?“we install Hadoop in docker containers. And we run tests on 3 flavors of Hadoop (HDP, CDH, and IOP)”
neutral · production · mauvehaus · 2017-07-24 · source ↗ · wrong label?“We had a cluster for Hadoop experiments at uni and no ressources to replace all the faulty disks at that time”
neutral · personal · nisa · 2017-03-13 · source ↗ · wrong label?“We use Hadoop-like stuff on, ahem, sizable quantities of data.”
neutral · production · kstrauser · 2016-12-02 · source ↗ · wrong label?“We run many different types of data services, mysql/hadoop/hbase/kafka/redis/zookeeper/elasticsearch/etc”
neutral · production · paulfurtado · 2016-11-04 · source ↗ · wrong label?“We ran at Uni a Hadoop Cluster that had disks slowly dying (some bad sectors every few days, but otherwise fine)”
neutral · production · nisa · 2016-08-07 · source ↗ · wrong label?“we ported dremel to our heavily customized hadoop (it is, even now, at least 10x more performant than the public hadoop code”
neutral · production · x0x0 · 2015-10-28 · source ↗ · wrong label?“VPN access to the hadoop clusters I work with.”
neutral · production · francisb07 · 2015-07-27 · source ↗ · wrong label?“I work on Hadoop clusters nowadays and any time we run into a load of 100+ there is a severe degradation in the service”
neutral · production · istvan__ · 2015-06-19 · source ↗ · wrong label?“Last time I tinkered with Spark (this was over a year ago) it was so buggy, next to useless”
neutral · personal · gtrubetskoy · 2014-10-10 · source ↗ · wrong label?“processing raw user behavioural data at scale using MapReduce.”
neutral · production · alexatkeplar · 2013-11-10 · source ↗ · wrong label?“We use our own highly customized fork of Hadoop to generate traffic graphs [1] and demographic information”
neutral · production · Tossrock · 2013-11-10 · source ↗ · wrong label?“We use MR using Pig (data in cassandra/CFS) with a 6 node hadoop cluster to process timeseries data.”
neutral · production · zengr · 2013-11-10 · source ↗ · wrong label?“we run Hadoop on top of CephFS, and can deal with the occasional metadata server problem.”
neutral · production · noahdesu · 2013-05-05 · source ↗ · wrong label?“we already some Hadoop projects”
neutral · production · linuxhansl · 2011-11-07 · source ↗ · wrong label?“We use Java, MongoDB, Hadoop and embedded Jetty at MetaBroadcast”
neutral · production · johnayres · 2010-11-23 · source ↗ · wrong label?“we run on the "web scale"... It's why we're built on Hadoop.”
neutral · production · gsteph22 · 2010-01-25 · source ↗ · wrong label?