Bestand wählen
Merken

Infrastructure design patterns with Python, Buildbot, and Linux Containers

Zitierlink des Filmsegments
Embed Code

Automatisierte Medienanalyse

Beta
Erkannte Entitäten
Sprachtranskript
are good afternoon everybody my name's david blue and I am the Python technical consultant
engineer at Intel and so normally a lot of my focuses on the numerical American scientific side but occasionally I you know I I have a a big my work in infrastructure so you know on that topic I've come up with some interesting ways of using build button so I thought I'd give you a chat on how hard than so just kind of is an overview of Bernhard just talk about it but what you can do and what I mean by infrastructure design patterns what build body as a more to talk about them looking things up in unusual ways of port simultane masters and pseudo remote procedure calls and then drop into when use containers putting all together with Python and then talking about some more like light examples and things that I've actually implemented and in the real world so just on the term of infrastructure no this term is thrown around a lot and what exactly is automated uh was what exactly is infrastructure is that automation is orchestration reading in task distributed task runners I think there's a lot of confusion as to what is actually is an in many options actually exists in in the Python world for what you may need so and you can have desk at i per alone jobble before more distributed task runners for numerical work on you have orchestration the chef and puppet under cellaring Kafka which are also also heavily tied in with Python that are more automated task and scheduler runners so many of these examples are very heavy handed so like they'll do a lot but you have a lot of set up because they're not necessarily meant for this specific task so uh you usually have 1 solution that's way way too too complicated for that job that you're actually using it for the so some examples of 1 of these really odd things words too heavy handed it would be a distributed task system such as task which I really really love to run a con job which is completely out of the scope of what it's supposed to be doing and as uh not a good use case trend accelerated you mass a MapReduce operation which should generally doesn't like to do and are trying to get public to make a task graph so these are all examples of when you basically have and computing features that don't have that don't exactly fit under the paradigm whatever task doing so you're using these frameworks against what they were designed for so to get out of that mentality of using you just as big pieces just kind of force them into a certain direction of what you want at the way that i've approaches to take other frameworks that a loose enough that you can use as building blocks to get that so 1 of the 1 that I use is built by which is normally used for continued integrations written in Python uses twisted as as as back and and and pretty popular because it builds the actual Python language I think if you the languages like rust and you can construct elements of in really weird ways so to sec Lago block you can just take pieces bits and pieces and arrange them in an odd order to get the job done so continuous integration the CI tests generally you know incur it incorporate a lot of interesting pieces you have a scheduler you have dependencies and usually have sensible result at the end that you're trying to achieve and but the main task apologetic composed of other primitives that you can break up into smaller pieces and utilize if you choose the right and the right pieces so some of these the examples of these pieces are what you see up here on some of the more notable ones are was like triggers resource calls and distributors communications the scheduler itself on and build steps so a lot of these you probably seen if you've ever done any type of continuous integration or build and so the when you look at these components you know in your head you think OK this is meant for this is meant for this but when get a hopefully show you today is that you can do a lot more with the components and demonstrating appear and so 1 of the things that is unusual about Bill but is that because all the things I split up they give you the option to wire them in usual ways and that's when they reveal the upper layer to you normally like for Jenkins a team city OK I'm just using it in this in this format but if you start to break up the pieces and say well what the schedule reporting to another sketchily the reporting to something else then you you achieve some design flexibility on so before I go on to there's a huge huge warning as someone who is formally in like a security field need the examples I'm about to show implement 0 security and before you ever put this in production or or anywhere internally just understand that there is very very little security or orchestration is going on so just just understand that before and this is considered a very often use build by so we don't go you know asking the robot people hey how can I set it up for this can figure this really weird off the wall configuration is probably not supported use case so and the 1st steps in breaking out of and just the CIA mind-set would be that the the design patterns are most you take the most common tasks and roles in interconnects that that the suffer deployment so you take a look and say well what type of roles are always there in build is again just 1 way of of solving this so if you use if you use it you can't protect something and about something else when you put a production so some examples of what I've actually implemented in well show you as an example I've I've listed appear on enterprise application deployment and license management is has been to the most popular uses of that i've that i've may prove concepts forcible go into that in a sense but there's other things that happened on before but so talk about where you where you actually need to start doing this so bill that has a lot of interesting components when you need to actually work with us with a with a worker and a master and 1 of the things it exposes ports but you can use the way that it exposes a in a non-standard way so what I what I actually do is change port in in Bilbao allows you to normally say well I want to trigger built from an external source but what I'm going to show you is that you can use a symlink all to a Python script which then gives you a kind of a user triggered to trigger a whatever you want to give you that very much alike like if you map something user been to it you can actually watch an application so and bypassing argument agonists of those on 3 that Python I can also kind of remote procedure call into the worker because the workers just interpreting it as Python and the port through the through the password so that you can actually inject Python code again you know sense sizes inputs here but you can do some really weird things with that and and most that logic is control to the master doctor and CFG which is in the majority interpreted as Python code so here's an example of what it actually looks like in the pi bond and yet CFG you normally have a change source right and and you usually ask it to pull from a repository on Poland's repository all the time OK as great you know some some check something it'll it'll update but then you can actually add a secondary change source and that's the method that I'm about to show you is I'm adding that secondary change 1st uniport and then injecting commands into it the so 1 of the files that you can use had is the fake change that PY that a ship in the by contribute on and what it does is it accesses that port and sends that actual build command to the specific keyword project so if you create the scheduler for it and you create the the change request for it a match the 2 together then you can send a change the schedule and then you can trigger whatever you want be in an application be it on you know as a task of sometimes so I I
know this is that a lot of talking and as public to see it in action so here's kind of an example of a running so in this
case I've actually symlinks and some some commands to the Linux sorry to the Python script and I'm calling it it's giving me a result back
and in this case is actually running on 1 of these uh what I would consider it Linux issue shell in Dr. for on a
on a different worker so in this case from 1 space and actually triggered the ability to
land a Linux session hosts aware that you know in a in a big computer and right from time just just this simple pattern of injecting that code all the way into their and then asking it to then build what in the build I'm calling and that containerized application to then run to the user in that example that I show that just flash before you I showed the actual on project command and that was being used in which I think was right here so I was calling it to then say I want you to build
this project actually that project is a bunch of Python code that's been calling the dock replication and next you can
think outside of uh just to the standard build process you can start going to multi-master and which creates some really really you all situation so you can go from uh 1 worker than being load balanced to another worker through like you know you the multi-master set up or a resourceful like situations so but you know 1 of the things that I like to say is don't hesitate to kick off another 1 of the of such as a bill bill but instances you can have
distress that you created an abstraction that just use that mostly works and you can just kind of play around in that space but occasionally you will have to break out of just a single scheduler or master to achieve the task at hand
and to also expand upon what I just said what type of command and actually injecting into and the bill but worker but is this example so in this case I I am hard setting this this display rules you can see what if if someone's display was passed as an argument during that during that from fake change script over you then assigned that's too and that the command that you're calling on the worker and then you will be able the call the entire document in this case the top 1 is for Emacs in the bottom 1 is for the the the kind of Richard terminal that I was showing a 2nd ago so again I can have these in a mix even on the same worker with the same build script because uh you're essentially tying each of the commands to a different project of light bulb out will consider a buildable project so the reason that I'm even showing parts of Linux Containers is that they have the advantage of being a great abstractions to things that don't normally so when something do it doesn't want fit together or you have a budget dependencies you can actually use the Linux Containers systems are put together to actually equal a task and you can you know you do that 2 doctors you do that through uh any any like rocket fire any of this the the classic my container technologies and in the case I was shown as Rt showing the doctor with with clear containers underneath that prevented from getting out of the KVM uh escalation so the bill with with containers typically what you wanna do is you want a cordon off the risk your bet so if you don't want things escalating out or you don't want uses mapping things to the volume when you using the container then our the application in this case then this is 1 way of going and another thing that you can actually use this bright privileged and non-privileged barriers to separate users on so you can make it make sure that you know there's a uh that that's an application that's very dangerous to something can't be used for about 10 can escalate out into something else and then at some point you you will you may need at orchestrations book tasks so just no responsibility what what technologies and if any other problem and you it depends orchestration could you could muster the situations I've never had to actually use orchestration tuple something offices and I've seen some dangerous things puppet on by if you need it Erica organization use that and something consider and it like what I was saying before false fails just bundle everything up in a in a container and automate that as a single component so you know in the in the concept of pipe and the reason that this does have a critical component in Python is that the code bigger passing from even in your um script level application that I was showing the fake change that was linked you can pass in Python code gets passed in the master which then gets passed in all the way down to the worker so you're essentially go copying that code from 2 or 3 machines were with the resources are a so very specific build of Python gonna number pi you can pull it off pretty easily onto to to get that command over and Python is as we know 1 of the best glue code uh capabilities in infrastructure so that's another reason that I'm I think has been a very and critical piece in a lot of the stuff that I designed so i'm gonna show off to things and to show off the power company wide so server application deployment works and and that's actually again using the same mechanism which is similar in Python scripts and then we have the license server for a floating license so win if you have say an application that has a floating license but it can get out because of the uh like of a a company proxy a firewall then this is 1 way that you can actually balance that license without having contention aren't as the given application resource and so the way that I
designed this 1st example is that but you know you have something user been that's not that's mapped to the port change that I just demonstrated and that could talks to another server and so not not not like to use a computer the log node you talking some with full privileges and it's running the application for you and handing back on the X. forded screened which becomes your application screen and you can update the applications based on the back and through like any type of repository system in this case and doing through like dark repository and stuff so now and then you can spit new workers if you need more instances so that's kind of where i've heard orchestration comes in handy is like to do so the load balancing with that is you to text so In this case what
I'm gonna show you is how I set set this entire
example up and and the 1st thing I do is I run the I have a doctor composed of the 3 containers that are running both that the master and 1 of the builders this I have I have both a master builder and then I have to workers that are that all individually start up but in this case all start showing you and with the worker looks like in the fact that Bill by does come with a Kuwaiti so when you wanna look at the dashboard you can actually um see what's going on in interact irectly with it's a few users if they're doing stuff that requires a direct application what might be
useful I mean in this case I'm going to and it's it's local
but it's local page and you can look at with a which which 1 of your workers are up which task can be accomplished at which workers on the concept within billboard bodies that you tied very much the resources and capabilities of given worker to a specific job so in some cases maybe 1 worker can do to the jobs we are working 2 3 jobs depending on what application and you want for and so in this case you can actually see that i've started up but in but the workers are offline and so really I can't do anything until the workers come up and so you can simulate that is saying well I don't have any privilege resources of those were those application servers actually running the and so you know I have to go start as a so in this next part
I'm manganese going to start start
this up and notice how dangerous is X X has pluses and this good point that out here on I'm starting at each of the workers
and you'll notice that all connect to the uh to the master instance and so in this in this case they're communicating with each other and hopefully when you look at the the main page I think I screwed up over here somewhere yes I think that when you look at the main page you'll see that both of them the workers and green and they're showing
that they're capable and ready to work and I
think so you can see the 2 names of the workers so say you have a given application that can only run with that particular machine are more hasn't you know given processor or mechanical so our our computer set up and yes you would need
that and then I can actually start the job
through this you do the change for a conserve and in this case I'm running Emacs and
you know you say OK that's cool that's just the Maxwell actually this is running set of the darker container that's then being controlled from and that the the actual bill worker so its ability to manage these is really useful when when trying to you know load balancer applications are prevent or or be about being able to
administer your applications on if you're an IT and administrator of some type so see so the next 1 of
hopefully we show a little bit more information about what you can actually see from this side to see your application you can see who ran it for how long and you can see how many times it was running who's
currently running at how many things are being used on so I think those all those are really important and again you can see the and the Cimlinc call so in this case instead of calling it directly from the uh the Curie um from the dashboard and then going to call it from the same way through the uh through the command line that's been some linked all the way through and again you can see start up and as
running on and you see a current state so in terms of uh being able to prove make for for concepts very quickly with this it's it's super useful when you want see what's going on in the hood especially with the uh with the
dashboard but again I wanna stress that this is really really just super useful for
concepts in that you know maybe this would be
better if you my greatest to something else as
and you figure out what your needs are as you go into production system so again you can see that these are 2 completely different and darker instances of Emacs running around and as we
close them you can see you can start seeing them uh cordoned off and everything call
so in terms of a floating license right so it it you don't have to do too much different in that previous case now do a floating license so on the licensed easy to be held by the by the database that's attached to the bill bot-master instance or you can use some type of uh the actual build logic within your build scripts to accomplish that so again it doesn't seem to differ from the previous example and then I'll hopefully show it
here the thank
you not of my my artifacts for this this time OK so I'm gonna force 1 of and to start immediately and and so let's just
assume I have to um pools
are I guess I have 1 pool with 2 with 2 slots right so instead of those being machines there now instances could you can
uh start out the bill but workers
on the on your when so maybe you have a machine that has 50 these workers and that becomes
your your resource pool right and so now I have 1 started up and down the to try to start up a 2nd all of them could see might have the fast forward through that of talking mentally article
so once the 2nd 1 starts at time you'll you notice that both both of the workers have now been taken up
it would and so right now the
the flashing yellow because the book being taken up so if this
was your resource polar this was your on your license the matter licenses that you have for the given application you can then block the next time instant you to the next person that
actually accesses in them when that other person
close at the next person in line gets and so in terms of queueing for the
users and this is really useful in kind of an enterprise Edition environment where maybe someone says OK they're going to use an application until want understand Q myself on line to come back and hopefully of landed might my session by then right
so unified effort take 1 out I can then but 2 1 which I just showed and then the next will start up and now both people again the next person in line has it so it you can use again the thought of this pattern what what exactly here's abstract the workers are abstract as either pieces of the resource pool or they can be used as another part of the whole as a license server and more how many licenses you have or you can spin them upon demand-based pond load so but the
along the types of things that I've actually done in industry to you know creates a like a proof of concept art give and maybe a small company 1 way of actually creating their their own internal infrastructure is by using something like this we done it I've done in a week uh session compute session Linux handler and so you can then use X X 11 14 like I just showed to pull that off and actually have like a whole machine learnings server which I used to kick off jobs in a costly computes geographic data as as it updated um and
so this to go back to the tree example of 1 that worked and the computer was a little more can um uh a little more complicated in that 1 the 1 that I showed in the video so and we did have a login know that we had a model that was a smaller machine that smaller processes and we had 1 that was a large server farm and we use those that had full of and the full resources and privilege to have their and with the workers and then with people bot-master and these session you keep those for the people who were logging into that node and so with the landed sessions and we're able to scale this I think to maybe about I think the most i've ever seen was maybe 3 or 4 100 users in the this is on a very small set up like with a log node of maybe 1 machine so I think that you know in terms of proof of making a proof of concept for what you're infrastructure should like it should be like then you know it would be a useful that point and then as you try to on scale up to something else you can look at other other types of infrastructure components to pull that
off Machine learning server that implemented use utilizes more of the continues integration components and so and achievable bot-master again with the session q the worker that's basically running all the time actually polls from aligned data source and then when things have updated correctly make gets enough aggregate data and then send it off to the compute workers and then post that back on my screen and that's all automated on my like you know right above my culture this monitor that runs all the stuff they can see with the you know the aggregate data for the for that thing as so again this is another utilization of the components but for completely different purpose so in
summary some you with a little bit of ingenuity and creative use components you can really pull together a lot of infrastructure design patterns that appear in like sort for IT with creative you know freedoms because as more of of these weird frameworks to technologies get released you're your your ability to create new and exciting uh if structure patterns just get get better so not being able to rapidly design these things with with this method allows you to look inherent problems in maybe the way you set things up before you go to a full-scale on the set up and I also like to state that remedy examples here natural with any security orchestrations be very wary of that and then you know that 1 of the things that you can do also just keep an eye on the upcoming technologies that show what I remember trying to do this before the days of you know really trendy Linux containers and it was pretty difficult in some instances and the that's open the doors of other things the orchestration technologies open up another door of things you can possibly choose to do and then you can experiment with new tools often see what happens to be made next it's
just the but and questions so
thank you so yeah I no question Russians OK we have a question yeah but do you have any experience with the good what travels which is providing a service of it will follow the integration group but not know but I have started looking at experimenting with that I was really excited when I want to make some of my own open-source projects build with that I'm and I have a really got the chance to experiment with that but I I do look for to experimenting with that the question this 1 year great talk things you mentioned a lot of these the proof of concept in security needs to work on on are you going to pursue that all to give people tips on how they could secure there's that kind of looked at the user to figure out I I mean I think with how different on a lot of infrastructures are with different companies and you know this is more in the domain of I. T. I. T. security which is less of an area that I was wasn't and I would really just I don't have that much that many tips to give I just know that I've left a lot of holes open and I'd like to trust and you know my my network IT guys in insecurity guys about what they what they think should be done in this case the more questions and you may have it
RPC
Kontextfreie Grammatik
Aggregatzustand
Information
Computeranimation
Richtung
Intel
Code
Computersicherheit
Skript <Programm>
Softwaretest
Computersicherheit
Gebäude <Mathematik>
Güte der Anpassung
Stichprobe
Ruhmasse
Biprodukt
Scheduling
Software
Forcing
Rechter Winkel
Ordnung <Mathematik>
Lesen <Datenverarbeitung>
Mathematische Logik
Multiplikation
Mathematisierung
Mathematische Logik
Unternehmensarchitektur
Open Source
Quellcode
Datentyp
Widget
Skript <Programm>
Primitive <Informatik>
Gruppoid
Konfigurationsraum
Normalvektor
Architektur <Informatik>
Softwarewerkzeug
Interpretierer
Elektronische Publikation
Gamecontroller
Wort <Informatik>
Quadratzahl
Resultante
Bit
Konfiguration <Informatik>
Reihenfolgeproblem
Formale Sprache
Kartesische Koordinaten
Computerunterstütztes Verfahren
Kardinalzahl
Kontinuierliche Integration
Eins
Komponente <Software>
Datenmanagement
RPC
Prozess <Informatik>
Gamecontroller
Programmierparadigma
Umwandlungsenthalpie
Parametersystem
Nichtlinearer Operator
Softwareentwickler
Dokumentenserver
Entwurfsmuster
Element <Gruppentheorie>
Systemaufruf
Prozessautomation
Übergang
Quellcode
p-Block
Ein-Ausgabe
Systemaufruf
Konfiguration <Informatik>
Datenfeld
Twitter <Softwareplattform>
Framework <Informatik>
COM
Projektive Ebene
p-Block
Versionsverwaltung
Beweistheorie
Prototyping
Telekommunikation
Server
EDV-Beratung
Term
Code
Framework <Informatik>
Task
Systemprogrammierung
Physikalisches System
Graph
Task
Zusammenhängender Graph
Schreib-Lese-Kopf
Matching <Graphentheorie>
Graph
Kontinuierliche Integration
Mathematisierung
Physikalisches System
Integral
Roboter
Injektivität
Unternehmensarchitektur
Brennen <Datenverarbeitung>
Resultante
Nabel <Mathematik>
Gruppenoperation
Skript <Programm>
Computeranimation
Flash-Speicher
Rechter Winkel
Mustersprache
Kartesische Koordinaten
Projektive Ebene
Computer
Kontextbezogenes System
Code
Raum-Zeit
Computeranimation
Teilmenge
Multiplikation
Task
Prozess <Physik>
RPC
Last
Datenreplikation
Projektive Ebene
Code
Computeranimation
Kette <Mathematik>
Instantiierung
Punkt
RPC
Datensichtgerät
Kartesische Koordinaten
Raum-Zeit
Computeranimation
Übergang
Code
Minimum
Radikal <Mathematik>
Mixed Reality
Freier Ladungsträger
Skript <Programm>
Punkt
Feuchteleitung
Kette <Mathematik>
Kraftfahrzeugmechatroniker
Parametersystem
Abstraktionsebene
Systemaufruf
Dateiformat
Systemaufruf
Hoax
Scheduling
Software
Menge
Server
Projektive Ebene
p-Block
Message-Passing
Proxy Server
Server
Subtraktion
Multiplikation
Selbst organisierendes System
Firewall
Mathematisierung
n-Tupel
Zahlenbereich
Gebäude <Mathematik>
Supercomputer
Code
Spezifisches Volumen
Task
Virtuelle Maschine
Service provider
Task
Endogene Variable
Datentyp
Skript <Programm>
Zusammenhängender Graph
Inhalt <Mathematik>
Spezifisches Volumen
Leistung <Physik>
Architektur <Informatik>
Einfache Genauigkeit
Schlussregel
Physikalisches System
Office-Paket
Mereologie
Data Mining
Server
Dokumentenserver
Dokumentenserver
Mathematisierung
Kartesische Koordinaten
Physikalisches System
Computer
Computeranimation
Warteschlange
Knotenmenge
Last
Datentyp
Server
Cloud Computing
Versionsverwaltung
Touchscreen
Instantiierung
Server
Stellenring
Interaktives Fernsehen
Kartesische Koordinaten
Computeranimation
Task
Server
Prozess <Informatik>
Mereologie
Server
Kartesische Koordinaten
Benutzeroberfläche
Computeranimation
Homepage
Server
Punkt
Green-Funktion
Computeranimation
Homepage
Instantiierung
Virtuelle Maschine
Server
Kartesische Koordinaten
Computer
E-Mail
Computeranimation
Server
Menge
Prozess <Informatik>
Mathematisierung
Kartesische Koordinaten
Computeranimation
Lastteilung
Server
Bit
Datentyp
Systemverwaltung
Kartesische Koordinaten
Information
Computeranimation
Server
Vererbungshierarchie
Systemaufruf
Term
Aggregatzustand
Server
Physikalisches System
Biprodukt
Normalspannung
Computeranimation
Instantiierung
Server
Datenhaltung
Systemaufruf
Term
Mathematische Logik
Computeranimation
Rechter Winkel
Touchscreen
Login
Datentyp
Skript <Programm>
p-Block
Bestimmtheitsmaß
Instantiierung
Virtuelle Maschine
Server
Forcing
Computeranimation
Instantiierung
Virtuelle Maschine
Server
Rechter Winkel
Computeranimation
Server
Kartesische Koordinaten
Computeranimation
Server
Last
Rechter Winkel
Mustersprache
Mereologie
Server
Kartesische Koordinaten
Unternehmensarchitektur
Programmierumgebung
Term
Gerade
Computeranimation
Server
Prozess <Physik>
Punkt
Reihenfolgeproblem
Computer
Computer
Computerunterstütztes Verfahren
Term
Unternehmensmodell
Login
Computeranimation
Videokonferenz
Netzwerktopologie
Multi-Tier-Architektur
Virtuelle Maschine
Physikalisches System
Knotenmenge
Prozess <Informatik>
Datentyp
Skript <Programm>
Zusammenhängender Graph
Algorithmische Lerntheorie
Zentrische Streckung
Architektur <Informatik>
Beweistheorie
Server
Bit
Server
Computer
Framework <Informatik>
Computeranimation
Virtuelle Maschine
Komponente <Software>
Mustersprache
Computersicherheit
Zusammenhängender Graph
Datenstruktur
Touchscreen
Computersicherheit
Softwarewerkzeug
Entwurfsmuster
Stichprobe
Quellcode
Quick-Sort
Integral
Software
Whiteboard
Server
Beweistheorie
Aggregatzustand
Instantiierung
Dienst <Informatik>
Domain-Name
Datennetz
Flächeninhalt
Offene Menge
Computersicherheit
Beweistheorie
Gruppenkeim
Projektive Ebene
Computeranimation
Integral

Metadaten

Formale Metadaten

Titel Infrastructure design patterns with Python, Buildbot, and Linux Containers
Serientitel EuroPython 2017
Autor Liu, David
Lizenz CC-Namensnennung - keine kommerzielle Nutzung - Weitergabe unter gleichen Bedingungen 3.0 Unported:
Sie dürfen das Werk bzw. den Inhalt zu jedem legalen und nicht-kommerziellen Zweck nutzen, verändern und in unveränderter oder veränderter Form vervielfältigen, verbreiten und öffentlich zugänglich machen, sofern Sie den Namen des Autors/Rechteinhabers in der von ihm festgelegten Weise nennen und das Werk bzw. diesen Inhalt auch in veränderter Form nur unter den Bedingungen dieser Lizenz weitergeben
DOI 10.5446/33698
Herausgeber EuroPython
Erscheinungsjahr 2017
Sprache Englisch

Inhaltliche Metadaten

Fachgebiet Informatik
Abstract Infrastructure design patterns with Python, Buildbot, and Linux Containers [EuroPython 2017 - Talk - 2017-07-12 - PyCharm Room] [Rimini, Italy] In today’s world of fast-paced development, infrastructure can get left behind quickly, leading to a potential increase in technical debt. Buildbot is normally known to be a continuous integration (CI) framework built in Python, but can be refashioned to solve infrastructure design patterns that arise in enterprise or production and deployment situations. Using Python and native Buildbot components paired with Linux Containers, patterns such as license management, resource allocation, load balancing, and enterprise application deployment can be architected quickly with room for expansion as one’s needs grow. Learn how to move past the CI mindset and construct infrastructure needs with Buildbot and popular Linux Containers such as Docker and ClearContainers. Attendees will learn the best known methods of configuring Buildbot in non-CI implementations, and how to utilize the framework components for future needs

Ähnliche Filme

Loading...
Feedback