Hi Fastly customers + friends,

In Q4 2019, we ended the year with a bang by launching Fastly TLS and Concierge TLS into General Availability (GA). We also announced Compute@Edge, the next step in Fastly’s edge computing capabilities and performed a demo at our customer summit, Altitude NYC. In addition to our exciting new launches, we continually extend our network, enhance our Fastly control panel, and add to our Varnish stack. For more information on recent releases, continue reading below.

Table of Contents

Network Updates: 68 global POPs, 74 Tbps of edge capacity

We now operate 68 POPs worldwide and recently deployed sites in Dublin, London, and Vienna. Our new sites and upgraded POPs extend our total connected edge capacity to 74 Tbps as of December 31, 2019.

Network Map DEC 2019

Fastly TLS and Concierge TLS GA

Fastly TLS and Concierge TLS went GA on November 18th, 2019.
Fastly TLS offers transparent, affordable pricing, with five free TLS enabled domains per account. It automates the entire certificate management lifecycle — from acquisition to renewals, and from upload to expiry notification. Built on Fastly’s powerful edge platform, end users will have seamless secure connections across Fastly’s global network without sacrificing performance and speed.

Concierge TLS is an add-on service aimed at customers with more complex TLS requirements or those needing more advanced configurations. It is available as an add-on to the Enterprise Support package.

Compute@Edge Beta

On November 6, we announced the beta product phase of our new serverless compute environment, Compute@Edge. Compute@Edge is a language-agnostic compute environment that offers 100x faster startup times than any current solution on the market and empowers developers to build sophisticated applications at the edge. This new level of edge computing capability elevates stability and security at a global scale.

During our annual customer summit, Altitude NYC, our Chief Architect Sean Leach gave a demo of Compute@Edge, showing a manifest manipulation use case in action. If you are interested in learning more about Compute@Edge, be sure to sign up for the beta program to be the first to know about future product news and updates.

Fastly control panel enhancements

  • Updated ‘Choose your logging endpoint’ user interface
    As part of our ongoing effort to simplify our self-service offerings, we updated the 'Choose your logging endpoint' user interface. The new design adopts a cleaner, simpler approach to endpoint selection. The updated layout replaces logos with more recognizable text names for the logging products and improves scannability. This card style also provides a more consistent experience with the interface style used in the existing UI.

  • Updates to DataDog logging method and documentation
    Our Log streaming: Datadog customer documentation has been updated to reflect Datadogs distinct US and EU datacenter locations. Customers can now find instructions on how to set up Datadog for the EU inline with US instructions.

  • Sorting on All Services
    Customers with a large number of services can now easily sort their All Services page by stars, service name, type of service, and requests per second.

Network enhancements

On November 21, Fastly announced traffic delivery and reliability improvements to supplement standard protocols such as Border Gateway Protocol (BGP). We created a proprietary mechanism which allows us to proactively respond to early signs of path degradation on our end user connections. Our failure-detection mechanism uses transport-layer metrics to infer the health of a network path and then automatically moves traffic to an alternate path if the current path is experiencing performance degradation. This capability paired with our routing architecture allows our edge cloud servers to make precise failover decisions that only target the individual degraded connections.

Varnish and VCL updates

Fastly deployed the following Varnish enhancements to the worldwide Fastly Edge Cloud fleet in Q4 2019.

  • Fastly increased the number of both IPv4 and IPv6 addresses that will be accepted and used from origin domain DNS resolution results. Previously, we only accepted, health checked, and routed origin requests to a maximum of 16 IPv4 and 16 IPv6 origin server IP addresses. We now support up to 100 addresses each.

  • We replaced the previous 501 status code response with the more accurate 416 response when a request is received for a byte range that starts past the end of the object size on services where segmented caching is enabled. The 416 status code response was already supported on services that did not have segmented caching enabled.

Sie möchten loslegen?

Setzen Sie sich mit uns in Verbindung oder erstellen Sie einen Account.