DevOpsDaysPortugal 2019 - Pedro Pessoa - How to shave off 100ms off your distributed cache by just shifting your containers

During this talk I will demonstrate how to shorten the time distance from the typical Redis setup to the end user. Reducing that latency typically results in faster, more responsive applications and therefore happier users. This revised architecture will make use of recently launched services which will frictionless run our containers (Redis k8s) closer to our users while keeping our primary data stores and code untouched. This talk will also cover running Redis on containers (k8s) and the benefits of that portability.

Key takeaways:

Redis is easily portable to almost anywhere in your architecture
New services allow us to deploy Redis closer to our users than ever.

Transcript

1. Devopsdays Portugal June 4, 2019
2. The container THE MAIN EVENT https://github.com/stackpath/webdis/blob/master/Dockerfile 2
3. The container THE MAIN EVENT https://github.com/stackpath/webdis/blob/master/Dockerfile 3 curl http://<container address>/SET/myvalue/test {"SET":[true,"OK"]} curl http://<container address>/GET/myvalue {"GET":"test"}
4. Results the old way THE MAIN EVENT 4 Results the old way THE MAIN EVENT 4 - Singapore at 436ms.
5. Doing it differently THE MAIN EVENT 5
6. Ready, set... THE MAIN EVENT 6
7. Go! Less than 60s later: THE MAIN EVENT 7
8. What?! THE MAIN EVENT 8 - Singapore at 2ms!
9. Redis replication HOW 9 ssh root@151.139.51.6 root@151.139.51.6's password: Last login: Fri Mar 1 01:33:08 2019 from 108.161.176.6 root@webdis-north-america-iad-0:~# for i in 10.128.32.3 \ 10.128.64.2 10.128.96.2 10.128.128.2 10.128.0.3 \ 10.128.112.2 10.128.176.2 10.128.80.2 10.128.144.2 do echo $i redis-cli -h $i -p 6379 slaveof 10.128.160.2 6379 done 10.128.32.3 OK 10.128.64.2 OK 10.128.96.2 OK 10.128.128.2 OK 10.128.0.3 OK 10.128.112.2 OK
10. What is the Edge? DEFINING THE EDGE The edge is an end user or device’s first point of connection to the Internet. USER 1 USER 2 The Internet is, basically, data centers around the world connected by cables. A person or device connects to the closest data center and their data then hops from data center to data center to until it reaches the data center they are really trying to communicate with. The “edge” of the internet is that point of first connection. That means “edge” describes the role a facility plays in an end users’ traffic, not the type of facility. A specific data center (or internet exchange, network access point, cell phone tower, etc.) may be the edge for one user, but the same facility is a hop for other users. And it means that there are degrees of “edginess.” A facility might not be a person or device’s absolute first point of connection, but if it is two hops away in an eight hop path, it’s very close to the “edge.” The light orange data center is the “edge” for User 2, but it is not for User 1. ORIGIN ORIGIN Edge Edge 10
11. DATA CENTER/CLOUD EDGE IX/ENTERPRISE EDGE TELCO/CABLE EDGE END USER EDGE 50-100ms / Dozens 25-50ms / Hundreds 5-15ms / Thousands 1ms / Millions Network Response Time / No. Locations Where is the edge today? 11 DEFINING THE EDGE Our solutions currently reside in IXs worldwide but are designed to be deployed at any point of the edge. DLL EQX Container Data Center AWS Phone Data Internet Security IoT Mobile Device CyrusOne Telx Central Office Google Owned Interxion Data Hut Azure 5G LTE Broadband 5G Cable Fiber Central Office Extension Voice Backhaul Data Services
12. What is edge Computing? DEFINING THE EDGE Edge computing is processing data at the “edge” as opposed to the “origin.” Origin computing centralizes processing, storage, and end-user delivery in a limited number of larger-scale public cloud, collocated, or on-premises data center (“origins”). This requires additional costs to distribute data and manage the connectivity between networked locations. Edge computing shares processing, storage, and delivery responsibilities between origin data centers and a network of lighter-weight edge nodes located closer to end users. This allows significant and unique workload optimization, including higher security options, reduced bandwidth costs, accelerated processing time, and superior end-user experiences. 12
13. DEMO 13