Weekly update 11

Hi, wel­come to anoth­er week­ly update! Apolo­gies for the lack of update recent­ly, a com­bi­na­tion of back injury, busy work sched­ule and week long hol­i­day in Mal­dives have got in the way of updates. Good news is that I’m back, and I have a num­ber of guest posts in the pipeline that I think you’ll real­ly enjoy 

MEAP update for Production-Ready Serverless

We have released 4 new videos on my video course Pro­duc­tion-Ready Server­less!

Remem­ber, whilst the course is still in the MEAP, you can get 40% off the face price using the code ytcui.

I have also turned this course into a class­room set­ting, so if you’re inter­est­ed in hav­ing me run this as a work­shop in your com­pa­ny, then drop me a line via the Con­tact page.

New Posts

Best prac­tices for AWS Lamb­da time­outs. I wrote a guest post for Epsagon where we looked at ways to pro­tect the user expe­ri­ence dur­ing laten­cy spikes, and how to strike a good bal­ance between per­for­mance & cost. We dis­cussed why you should use short time­outs for your Lamb­da func­tions and we also looked at the time­out setting’s impact on secu­ri­ty and cold starts.

The 5 best use cas­es for the Server­less begin­ner. Anoth­er guest post for Epsagon where we looked at the 5 best use cas­es for peo­ple to start adopt­ing server­less tech­nolo­gies with — ops automa­tion, web apps, ana­lyt­ics pipelines, batch pro­cess­ing and IOT.

Future of Server­less: a toy that will evolve and offer flex­i­bil­i­ty. I spoke with Corey Quinn recent­ly on his pod­cast Scream­ing in the Cloud where we talked about AWS, server­less and chaos engi­neer­ing. If you haven’t sub­scribed to his newslet­ter Last Week in AWS then you’re miss­ing out!

Apply­ing prin­ci­ples of chaos engi­neer­ing to Server­less. I gave a talk at the recent SRE­Con in Ger­many on how we can apply prin­ci­ples of chaos engi­neer­ing to a server­less archi­tec­ture and test the resilience of our server­less appli­ca­tions.

How to build observ­abil­i­ty into a Server­less appli­ca­tion. I gave a talk at the recent Mon­i­tora­ma EU in Ams­ter­dam on how you can build observ­abil­i­ty into your server­less appli­ca­tion, includ­ing many of the lessons and tools that I have been build­ing and using at DAZN.

Upcoming Talks

As we approach the busy con­fer­ence sea­son, I have quite a num­ber of con­fer­ence talks com­ing up in Octo­ber:

  • CoDe-Conf (11th Oct, Copen­hagen): a one day con­fer­ence focused on Con­tin­u­ous Deliv­ery and DevOps
  • AWS User Group (17th Oct, Lon­don): a 2.5hr work­shop on how to build pro­duc­tion-ready server­less appli­ca­tions
  • DevSec­Con Lon­don (18–19th Oct, Lon­don): all about DevSec­Ops
  • Server­less­Days Milan (19th Oct, Milan): I’m hon­ored to be giv­ing the open­ing keynote for this year’s Server­less­Days Milan con­fer­ence 
  • Voxxed­Days Bris­tol (25th Oct, Bris­tol): a one day gen­er­al pur­pose con­fer­ence in Bris­tol
  • get.net (28th Oct, Gdan­sk): the biggest .Net con­fer­ence in North­ern Poland
  • O’Reilly Soft­ware Archi­tec­ture Con­fer­ence (29–31st Oct, Lon­don): O’Reilly’s annu­al soft­ware archi­tec­ture con­fer­ence where you learn from some of the best in the field
  • O’Reilly Veloc­i­ty (31st Oct-2nd Nov, Lon­don): O’Reilly’s annu­al con­fer­ence on dis­trib­uted sys­tems

The best thing about going to these con­fer­ences for me is to meet new peo­ple and find out all the cool stuff oth­er peo­ple are build­ing. So, if you’re attend­ing any of these con­fer­ences then grab me in the cor­ri­dors and say hi!

AWS Announcements

A new CLI tool­chain for AWS Ampli­fy to make it even eas­i­er to build AWS-pow­ered web and mobile appli­ca­tions.

Hor­i­zon­tal pod autoscal­ing now sup­ports EKS.

Cloud­For­ma­tion now sup­ports macros with Lamb­da!

Not an AWS announce­ment, but the server­less-step-func­tions plu­g­in has release v1.7.0 a new ver­sion with two changes I’ve want­ed for a while:

  • JSON pret­ty print the Step Func­tions def­i­n­i­tion
  • restrict the Lamb­da invo­ca­tion per­mis­sion of the auto-gen­er­at­ed IAM role to only the func­tions that are part of the state machine def­i­n­i­tion

Good Reads

Paul John­ston pub­lished a nice arti­cle on Server­less Best Prac­tices. It sum­marised many of the lessons that I have learnt over the last cou­ple of years while work­ing with Lamb­da, it’s well worth a read!

Nitzan Shapi­ra pub­lished a nice roundup of the most pop­u­lar deploy­ment tools out there for server­less.

Peter Pito pub­lished a nice arti­cle on how to use a Ward­ley map to build a roadmap.

Tero Parvi­ainen shared his slides for his talk on How gen­er­a­tive music work. I haven’t seen the talk myself, but the slides look amaz­ing, and I have long been a fan of gen­er­a­tive art so I was excit­ed to see sim­i­lar ideas applied to music.

And anoth­er beau­ty from Julia Evans, and pos­si­bly the most con­cise expla­na­tion for what a manager’s job should be:

Videos from Server­less­Conf SF are now avail­able online, put some time aside and watch as many as you can!

Chris Munns pub­lished part 2 of his series on how to pro­tect your API with API Gate­way and AWS WAF.

Mikhail Shilkov pub­lished an analy­sis of cold start per­for­mance across AWS Lamb­da, Google Cloud Func­tions and Azure Func­tions.

Dan North pub­lished an arti­cle on how we have mis­un­der­stood Monte Car­lo sim­u­la­tions.

Jere­my Daly pub­lished anoth­er good post on server­less secu­ri­ty, using Puresec’s new hot­ness — Func­tion­Shield!

Dani­lo Poc­cia pub­lished an arti­cle on server­less and evo­lu­tion­ary archi­tec­tures.

Mike For­tu­na pub­lished a tuto­r­i­al on how to visu­al­ize Guard­Du­ty find­ings, which looks pret­ty sweet!

And I’ll fin­ish off this week’s update with one of the best thing I have seen on Twit­ter for a while!