Tag Archives: xpds16

A Recap of the Xen Project Developer Summit 2016

The Xen Project descended on Toronto, Canada in late August for its annual Xen Project Developer Summit. The Summit is an opportunity for developers and software engineers to collaborate and discuss the latest advancements of the Xen Project software. It also gives developers a chance to better understand new trends and deployments in the community and from power enterprise users.

From community growth to new emerging use cases, the Summit covered a lot of ground. Developments within core technologies such as security, graphics support and hardware support were discussed. We also covered emerging technologies such as automotive, embedded and IoT. All sessions were recorded and are available here and also on slideshare (follow this link for summit presentations).

Below is a summary of a few videos that feature technology that has been recently introduced into the Xen Project hypervisor as well as emerging technologies that are being built with Xen Project technology.

New Feature Technologies from Xen Project Community and Power Users

In Xen Project 4.7, we introduced Live Patching as a technology preview. Live Patching gives system administrators and DevOps practitioners the ability to update the Xen Project hypervisor without the need for a reboot. Konrad Wilk, software development manager of Oracle and Ross Lagerwall, software enggineer at Citrix, provide insight into how it works, what the difficulties were to implement, and how it compares to other technologies for patching (kGraft, kPatch, kSplice, Linux hot-patching).



Dimitri Stilliadis, CEO of Aporeto, provides a great overview of the benefits of using Xen Project software to provide an execution environment for Docker apps. This approach allows VM-like isolations for security measures without having to sacrifice performance. The presentation introduces a new paravirtualized protocol to virtualise IP sockets and provides the design and implementation details.


Data breaches are happening all the time, and there are many ways that organisations are trying to stop this through detection, pattern matching and behavioural analysis. However, Neil Sikka, founder and CEO of A1LOGIC, provides a new way of looking at this problem and solving this problem by using the Xen Project hypervisor to enforce data loss prevention. It doesn’t use any type of detection, heuristics, pattern matching or behavioural analysis, but rather a strictly algorithmic approach rooted in hardware.

Embedded Projects and Xen Project Software

Members from the Xen Project sister community OpenXT, an open-source development toolkit for hardware-assisted security research and appliance integration, were present to provide some insights into how Xen Project is working within the embedded space and best practices for embedding Xen Project on mobile and tablet devices.

If this is an area that you are interested in, check out Christopher Clark (consultant and interoperability architect at BAE Systems) overview of the OpenXT Project, which has begun to attract new users and contributors. We also recommend Chris Patterson’s and Kyle Temkin’s step-by-step guide on the challenges and lessons to get Xen Project software started on phones and tablets. Chris is a advising computer engineer for AIS and Kyle is researcher for AIS.


Emerging Technologies

Xen Project is consistently becoming more common within automotive and aviation. Xen Project 4.7 introduced the ability to remove core Xen Hypervisor features at compile time via KCONFIG. This allows a more lightweight hypervisor, which is perfect for IoT scenarios and better for security-first environments, like automotive.

Sangyun Lee, senior embedded software engineer of LG Electronics, presents on the real-time GPU scheduling of XenGT in Automotive Embedded systems. It introduces the real-time GPU schedule of XenGT running on automotive embedded systems and explains why this should be used for an automotive system.


Xen Project is consistently being used within embedded systems for automotive. Earlier this year at CES, GlobalLogic showcased its technology behind Nautilus, which is the company’s virtualisation solution that enables multiple domains to share the GPU hardware with no more than a 5 percent overall in performance changes. More on this technology and how it uses Xen Project here.

The summit was a huge success with many interesting conversations. The Xen Project thanks everyone who attended and presented as well as the sponsors of the event Citrix, Huawei and Intel.

Trademark Policy Update

Attached is the final, complete Xen Trademark Policy as approved by the Xen Advisory Board and updated with your comments. Please feel free to offer any additional feedback you have as this document is meant to “live” as Xen continues to evolve.

Xen Trademark Policy June 08 in pdf

Also, the FIT discussed in the document is being worked on by the Advisory Board and I will publish the latest proposal shortly after it is reviewed by the Advisory Board.

Xen Trademark Policy Update

The updated Xen Trademark Policy based on your feedback is almost complete and ready for final community inspection; however, I did get a response from Citrix about the decision to restrict usage of the Xen name in products. Please feel free to add comments or send me any questions. Thanks.

The comment from Citrix legal is as follows:

Citrix, in conjunction with the Xen AB, seeks to protect the trademarks associated with the open source Xen® hypervisor in order to both strengthen the public’s identification of the Xen marks and prevent the widespread and unregulated use that can sometimes lead to trademark genericide.  This is for the benefit of all those in the Xen community who distribute and contribute to the open source project, and not just for the benefit of Citrix.  It is for these reasons that the Xen Trademark Policy was originally put forward.

With respect to the limited restriction on product names combining “Xen” with another name, Citrix believes this restriction is in keeping with XenSource, Inc’s longstanding use of “Xen” in the names of the company (“XenSource™”) and its flagship product (“XenServer™”), which is now marketed by Citrix.  Citrix believes, and hopes that the community understands, that use of other Xen-combined names might confuse potential users  of Citrix-sourced products as to the source of a particular product or service. Since all Xen-based commercial products on the market today (of which Citrix is aware) from other vendors are all non-Xen branded, Citrix believes that this is the appropriate time to clarify this issue.

For example, an ISV may create a service for registering servers running Xen and decide to call the service “XenRegister”.  A reasonable IT consumer could be confused and assume that the XenRegister service is sourced by XenSource and Citrix.  Instead, the ISV could call their service, for example, “VM Registration for the Xen® hypervisor” or “MegaRegister™ for Xen®,” or any other name which is in keeping with the Xen Trademark Policy and does not reasonably confuse an IT consumer as to its source.

Ultimately, the Xen Trademark Policy benefits all those who use or assist in the development of the Xen® hypervisor, in part by maintaining the value of the Xen Marks.  Citrix must police the marks in accordance with the Policy and in conjunction with the Xen AB in order for everyone to see value in their use as product identifiers.  Otherwise, the trademark can become a generic term (see escalator or aspirin in the U.S.) at which point anyone could call their hypervisor “xen” regardless of the source of the hypervisor.