I was recently looking at “sysdig monitor” (previously named sysdig cloud) and was impressed by this products list of features.
Deep container monitoring in realtime, peer deep inside containers to view latency stats, read logs and view process consumption via a web portal
View topology and interaction between containers
Docker monitoring, alerting + troubleshooting with intelligent Kubernetes, Mesos and Swarm integration
See what’s happening inside your k8s pods
Excellent tooling to monitor the mapping between microservices and highlight their dependency, latency and traffic transferred
Works with AWS, Google Cloud, GKE, Microsoft Azure, Rackspace, Digital Ocean, Softlayer
Replay performance history
Requires installation of a kernel module
Data is sent from the agent via HTTPS to the sysdig portal where you can view your application metrics
Integrates well with Azure Container Service
Collects JVM, JMX metrics from inside containers
Billing per month or year
You can use Docker Swarm Mode on Windows Server 2016 which requires the overlay network driver. Install patch KB4015217 to add in this functionality.
As part of KB4015217 HNS requires that IPv6 is enabled on Windows container hosts.
Snipped from the KB :
“Improved the Host Network Service (HNS) to support an overlay network driver for use on Windows Server 2016 to connect containers across hosts using Docker Engine in Swarm Mode.”
Getting Started with Swarm Mode
Windows Container Networking
Brendan Gregg explains how to troubleshoot performance issues with containers at DockerCon 2017.
Some in-depth troubleshooting tips here, excellent technical content, well worth watching more than once.
The CPU Bottleneck identification chart is very useful (slide 49) for methodology.
Here’s the slidedeck:
The tech press seems to of missed this public announcement from Microsoft. These will be the first Hyper-Threaded VM’s that will be offered via Azure.
Microsoft will be offering for the first time a new generation of Hyper-Threading Technology virtual machines for general purpose workloads, Dv3.
As well as a new Hyper-Threaded family for memory optimized workloads, Ev3.
There is no specific launch date indicated in the announcement however the announcement does mention that these will be matching the comparable AWS instance prices.
Just read a fantastic article on ScyllaDB from the folks at The New stack which probes into the pros and cons of this NoSQL wide column store.
What I like most about ScyllaDB is that it’s :
Wrote in C++ and not Java
Very low latency (we’re talking microseconds not milliseconds)
Requires no extra tuning unlike Apache Cassandra
Faaaast due to the shard-per-core design which avoids locks, uses own cache, wrote in C++ (no JVM)
API is compatible with Cassandra
Capable of 2 million requests per second per node
I would not use ScyllaDB just yet for production workloads but there is an Enterprise Edition coming out soon, all of Cassandra’s features will be implemented by the end of Q2 2017 and I believe that the user community will grow rapidly.
Definitely one to watch closely!
Some of the new features include :
Data dictionary A global data dictionary contains information about database objects in transactional tables rather than being stored in metadata files and nontransactional system tables
InnoDB Changes : Tablespace Encryption Providing data at-rest encryption for physical tablespace data files using Advanced Encryption Standard (AES)
InnoDB Changes : Partitioning Table partitioning is now from within the storage engine which can provide performance gains, but only under certain conditions and depends upon how your disks are configured
Account management Roles are now supported, which are simply a collection of privileges assigned to a user
https://dev.mysql.com/doc/refman/8.0/en/ Reference Manual
https://dev.mysql.com/doc/relnotes/mysql/8.0/en/ Release Notes
https://dev.mysql.com/doc/refman/8.0/en/mysql-nutshell.html What is new in a nutshell
https://dev.mysql.com/doc/refman/8.0/en/upgrading.html#upgrade-procedure-inplace Upgrade procedure