TwitterのIntent To Shipを眺めていたら発見した．Prototypeがリリースされるとのこと．以下のリンクのMotivationに導入の背景が説明されていた．
In order to increase privacy on the web, browser vendors are either planning or already shipping restrictions on cross-site tracking, such as phasing out third-party cookies. Third-party cookies are currently defined as those associated with a site that is different from the site of the top-level page. However, modern websites are typically served over multiple domains/sites, many of which are owned by the same organization. First-Party Sets provides a mechanism to group domains/sites belonging to the same organization as being same-party with each other, and thus defines a privacy boundary for websites.
When we first built AWS Lambda, we chose to use Linuxcontainers to isolate functions, and virtualization to isolatebetween customer accounts. In other words, multiple func-tions for the same customer would run inside a single VM,but workloads for different customers always run in differentVMs.
To illustrate this complexity, the popular combination of Linux Kernel Virtual Machine  (KVM) and QEMU clearly illustrates the complexity. QEMU contains>1.4million lines of code, and can require up to 270 unique syscalls  (more than any other package on Ubuntu Linux15.04). The KVM code in Linux adds another 120,000 lines. The NEMU  project aims to cut down QEMU by removing unused features, but appears to be inactive.
Firecracker’s approach to these problems is to use KVM(for reasons we discuss in section 3), but replace the VMM with a minimal implementation written in a safe language. Minimizing the feature set of the VMM helps reduce surface area, and controls the size of the TCB. Firecracker contains approximately 50k lines of Rust code (96%fewer lines than QEMU), including multiple levels of automated tests, and auto-generated bindings. Intel Cloud Hypervisor  takes a similar approach, (and indeed shares much code with Fire-cracker), while NEMU  aims to address these problems by cutting down QEMU.
The execution of the customer code happens on the Lambda worker fleet, but to improve cache locality, enable connection re-use and amortize the costs of moving and loading customer code, events for a single function are sticky-routed to as few workers as possible.