Mirage - Extreme specialization of cloud appliances (OSCON 2013)

Slide 1

Slide 1 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 1/36 Anil Madhavapeddy University of Cambridge Richard Mortier University of Nottingham Dave Scott Citrix Systems R&D MIRAGE: EXTREME SPECIALISATION OF CLOUD APPLIANCES @avsm @mort___ @mugofsoup http://openmirage.org http://decks.openmirage.org/oscon13/ http://www.youtube.com/watch?v=2Mx8Bd5JYyo

Slide 2

Slide 2 text

Slide 3

Slide 3 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 3/36 THE CLOUD THREAT MODEL Type-safety in the application layer defeats several external threats. Type-safety in the kernel will make all external I/O safe, but at what cost?

Slide 4

Slide 4 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 4/36 THE CLOUD THREAT MODEL Most attacks never come from within a guest, only from external traffic. So why all these privilege checks?

Slide 5

Slide 5 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 5/36 KEY DESIGN INSIGHTS The hypervisor gifts us a stable hardware interface. Cures the curse of library operating systems! Protocol-level compatibility between special-purpose cloud appliances, e.g. a webserver, a database VM.

Slide 6

Slide 6 text

Slide 7

Slide 7 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 7/36 CONTRIBUTIONS The unikernel approach to building single-purpose appliances Library OS + high level programming interface Single-address space layout Evaluation of these techniques using a functional programming language (OCaml) Benefits of type-safety need not damage performance Static typing + modules = high level manipulation Language extensions for systems programming in OCaml

Slide 8

Slide 8 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 8/36 CURRENT VIRTUAL APPLIANCES Compiler has to stop at userspace. Every level has a different API, calling convention, and privilege requirements.

Slide 9

Slide 9 text

Slide 10

Slide 10 text

Slide 11

Slide 11 text

Slide 12

Slide 12 text

Slide 13

Slide 13 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 13/36 MICROBENCHMARKS! Unikernel appliances are: Smaller, both in kLoC & image size Boot faster, at packet round-trip time timescales High performance Type-safe

Slide 14

Slide 14 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 14/36 APPLIANCE IMAGE SIZE Appliance Standard Build Dead Code Elimination DNS 0.449 MB 0.184 MB Web Server 0.674 MB 0.172 MB Openflow learning switch 0.393 MB 0.164 MB Openflow controller 0.392 MB 0.168 MB All configuration and data compiled into the image by the toolchain. Live migration is easy and fun :-)

Slide 15

Slide 15 text

Slide 16

Slide 16 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 16/36 HOW LARGE IS LARGE?

Slide 17

Slide 17 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 17/36 EVENT DRIVEN CO-THREADS Garbage collected heap management is more efficient in a single address-space environment. Thread latency can be reduced by eliminating multiple levels of scheduling.

Slide 18

Slide 18 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 18/36 THREAD SCALING SINGLE INSTANCE Threads are heap allocated values, so benefit from the faster garbage collection cycle in the Mirage Xen version, and the scheduler can be overridden by application-specific needs.

Slide 19

Slide 19 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 19/36 BLOCK STORAGE Additionally, reading from a Mirage NAS-like storage VM achieves 942MiB/s buffered, and 1.8GiB/s unbuffered.

Slide 20

Slide 20 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 20/36 TECHNIQUES Several implementation techniques give rise to these benefits: Simplified memory management Zero-copy IO buffer management Hypervisor security extension for VM sealing (W^X)

Slide 21

Slide 21 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 21/36 MEMORY MANAGEMENT SPECIALISED! Compiled native source code and runtime statically linked with random start offset. IO memory is mapped into a reserved area and can be distinguished. OCaml heap is contiguous, with simpler write barriers as a result.

Slide 22

Slide 22 text

Slide 23

Slide 23 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 23/36 OPTIONAL VM SEALING Single address-space and no dynamic loading W^X address space Address offsets are randomized at compile-time Dropping page table privileges: Added freeze hypercall called just before app starts Subsequent page table updates are rejected by Xen Exception for I/O mappings if they are non-exec and do not modify any existing mappings Very easy in unikernels due to focus on compile-time specialisation instead of run-time complexity

Slide 24

Slide 24 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 24/36 MACROBENCHMARKS We have implemented several larger appliances. We discuss deens, our DNS server in detail here. We also have: a simple webserver, an OpenFlow Switch, and an OpenFlow Controller.

Slide 25

Slide 25 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 25/36 DNS SERVER CODE let main () = lwt zones = read key "zones" "zone.db" in Net.Manager.bind (fun mgr dev -> let src = `any_addr, 53 in Dns.Server.listen dev src zones ) Cooperative threads as functions Statically evaluated configuration Functional callbacks Libraries directly link to network stack

Slide 26

Slide 26 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 26/36 DNS SERVER PERFORMANCE BASELINE Baseline figures, running Bind (standard) and NSD (high performance). NSD achieves around 70 kreqs/s.

Slide 27

Slide 27 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 27/36 DNS SERVER PERFORMANCE MIRAGE Comparing against Mirage appliance, with and without simple memoisation. This algorithmic optimisation added just 6 lines of code.

Slide 28

Slide 28 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 28/36 DNS SERVER PERFORMANCE C/MINIOS A rudimentary C-based appliance linking NSD directly against MiniOS. Poor user-space library performance vastly outweighs language effects.

Slide 29

Slide 29 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 29/36 SCALING VIA PARALLEL INSTANCES Request throughput for serving a simple static page using Apache on Linux vs. a Mirage appliance. Rather than pay the cost of interlocking for thread-level parallelism, we scale by running many instances of the Mirage appliance.

Slide 30

Slide 30 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 30/36 OPENFLOW CONTROLLER OpenFlow controller is competitive with NOX (C++), but much higher level. Applications can link directly against the switch to route their data.

Slide 31

Slide 31 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 31/36 SUMMARY OCaml is the baseline language for all new code Our C runtime is small, and getting smaller Is fully event-driven and non-preemptive Rewriting protocols wasn’t that hard Not necessarily the best research strategy though But an extremely useful learning experience Tech transfer is vital Unikernels fit perfectly on the cloud Internet protocol building blocks Seamless interop with legacy code through VMs

Slide 32

Slide 32 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 32/36 PURE OCAML CODE Device drivers netfront, blkfront xenstore Networking/Communication IP/TCP/UDP/DHCP/DNS(SEC) (v4) HTTP, SSH, OpenFlow (controller/switch) vchan IPC Storage NFS, FAT32 distributed k/v store 9P :-) GITHUB.COM/MIRAGE http://arakoon.org

Slide 33

Slide 33 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 33/36 MIRAGE ONLINE Website, Code, O'Reilly OCaml book, OPAM package manager Allows constraints to be applied to package installation Very useful for managing assembly of the many small OCaml modules that construct a Mirage appliance http://www.openmirage.org http://github.com/mirage http://realworldocaml.org http://opam.ocamlpro.com

Slide 34

Slide 34 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 34/36 KEY RESEARCH DIRECTIONS Interoperability — with billions of VMs out there A unikernel per-language? Interconnect strategies? Heap sharing? Formal method integration easier or harder? Coordination — planetary scale computers Resources are highly elastic now. How to coordinate a million microkernels? “Warehouse Scale Computing” Library Applications — where are they? Irminsule, a git-like functional distributed database Beanstalk, a self-scaling web server

Slide 35

Slide 35 text

7/31/13 Mirage: Extreme Specialisation of Cloud Appliances file://localhost/Users/mort/research/projects/mirage/src/v2/mirage-decks/files/slides/oscon13/complete.html#/ 35/36 WE'RE HIRING! MUST ENJOY OPEN SOURCE! OCamlLabs, Cambridge, UK Real world functional programming with OCaml Need compiler hackers, protocol heads, PL/type theory systems Networks-as-a-Service, University of Nottingham, UK (2 posts, 3 years, deadline Friday August 2nd) Joint with University of Cambridge & Imperial College Network virtualisation for millions of microkernel apps Need Mirage hackers, network protocol experts, graph theorists http://www.cl.cam.ac.uk/projects/ocamllabs http://bit.ly/13sBjjC