Monitorama PDX 2018, in Portland, offered an intense, three-day conference program -- by and for monitoring and DevOps practitioners.
You are here
What I Learned at AWS re:Invent 2017
One of the things I enjoy about working at Opsview is the opportunity to attend events like AWS re:Invent 2017 in Las Vegas. Amazon attracts over 40,000 attendees to re:Invent every year. Most of the customers with whom I work use some sort of Amazon's public infrastructure or it is in their plans. So this was a great opportunity to learn in-depth from industry experts.
To get acclimated to re:Invent, I explored the vendor expo on Monday. I spoke to a range of vendors, including some that are similar to Opsview. From the vendors to the conference itself, there were themes of predictive analytics, machine learning and how these technologies can be used today.
The keynotes by Andy Jassy, AWS CEO and Werner Vogels, CTO at Amazon, reinforced the machine learning message. Artificial Intelligence (AI) was discussed in the keynote, including a presentation from Brian Mathews, VP of Platform Engineering at Autodesk. Brian spoke about the impressive innovations Autodesk has made on AWS and how AI helps them complete workloads in the cloud.
Some of my session highlights:
GPS: Building a Profitable Next-Generation AWS MSP Practice
Speakers: Barbara Kessler, David Lim, Ben Perak
This session included great detail on how traditional MSPs are moving away from items such as device based SLAs and hardware based solutions to cloud/software based solutions and application based SLAs. The speakers also mentioned the opportunity for moving customers to the cloud and about partner enablement AWS offers.
Using Amazon CloudWatch for Amazon ECS Resource Monitoring at Scale
Speaker: Brendan McFarland, Mapbox
Brendan McFarland explained some of the challenges in monitoring a shared resource environment. These include tags at cluster level, multiple teams in one resource and loss of granularity for teams. His solution was to use custom metrics within ECS to grab information per cluster such as RunningTasksPerInstances and AgentConnectedPercent.
A Day in the Life of a Netflix Engineer III
Speaker: Dave Hahn, Netflix
Dave Hahn gave a packed audience a glimpse into the technology behind Netflix. He explained chaos engineering and how Netflix uses it to ensure highly available systems under a mutltitude of potential failure scenarios. If chaos engineering is new to you, I recommend taking a look at this free eBook.
Deep Dive on AWS CloudFormation
Speaker: Luis Colon, Anil Kumar
This session was useful to understand everyday management of CloudFormation. Luis and Anil explained StackSets and how they can be deployed into different regions and accounts. Their knowledge of use cases gave the session a sense of practicality for using CloudFormation. Dive further into StackSets.
Application Performance Management on AWS
Speaker: Marcos Ortiz
Marcos Ortiz covers a wide range of information including how to use CloudWatch X-Ray to APM best practices. I found this session quite insightful working in the monitoring space. Trends in monitoring hit home since I see customer environments are increasing in complexity. Check out this session recording for best practices on APM. Marcos includes a ton of links for resources to check out.
Auto Scaling Prime Time: Target Track Hits the Bullseye at Netflix
Speakers: Vadim Filanovsky, Anoop Kapoor, Tara Van Unen
The team at Netflix and Amazon explained a new product for Auto Scaling called Target Tracking. It allows you to autoscale based on an overall metric such as CPU utilization or throughput, for example. This leads to smoother scaling of your workloads across AWS by shifting the logic limits to the Auto Scaling system. Learn more.
AWS DeepLens Workshop: Building Computer Vision Applications
Amazon introduced a new deep learning camera for image recognition. The camera is designed to spread image-based machine learning to more people, including garnering interest from developers. The DeepLens cameras were given away for free as long as you signed up for a workshop. Everyone wanted to get in on this, myself included. AWS DeepLens can utilize frameworks such as Apache MXnet, Caffe and Tensorflow. Learn more.
The main feature of the camera is that it can use AWS SageMaker, which allows for you to train models at scale and optimize each of the 10 most common machine learning frameworks: https://aws.amazon.com/sagemaker/?nc2=h_a1
Working for a monitoring company, I find customers evaluating or already using Amazon's public cloud services. On the flight home to Boston, I found myself contemplating all the new ideas I'd taken in during the week. The highlight was getting hands-on with machine learning. Machine learning seems to be coming up a lot as something we all want to know more about. Besides the long lines and hectic shuttle buses, I can say I really enjoyed AWS re:Invent 2017.
More like this
Opsview's Bill Bauman and John Jainschigg attended Percona Live 2018 -- to talk about serverless computing, database monitoring, and catch up with...
Amazon Web Services is one of the fastest growing technologies being used by businesses to scale their...