Cloud Native Computing Foundation Cloud Native Rust Day EU 2021, 14 May 2021

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Apache Teaclave (incubating): An Open Source Universal Secure Computing Platform in... Mingshen Sun

Description

Don’t miss out! Join us at our upcoming event: KubeCon + CloudNativeCon North America 2021 in Los Angeles, CA from October 12-15. Learn more at https://kubecon.io The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy, and all of the other CNCF-hosted projects.

Apache Teaclave (incubating): An Open Source Universal Secure Computing Platform in Rust - Mingshen Sun, Baidu

Apache Teaclave (incubating) is a universal secure computing platform written in Rust, making computation on privacy-sensitive data on cloud safe and secure. Teaclave adopts multiple security technologies to enable secure computing, in particular, Teaclave is written in Rust to prevent memory-safety issues. It uses Intel SGX to serve the most security-sensitive tasks with hardware-based isolation, memory encryption and attestation. Teaclave is a function-as-a-service platform. With many built-in functions, it also supports a wide variety of tasks on sensitive data like privacy preserving machine learning, private set intersection, etc. Unlike traditional FaaS, Teaclave supports both general secure computing tasks and flexible single- and multi-party secure computation. In this talk, Mingshen will talk about the background, design and how Rust empowers Teaclave's implementation.

A

Hello: everyone, my name, is ming chun sun from baidu security on behalf of the tequila community. Today I will talk about our open source, universal, secure computing platform, written in rust, tcleaf I'll start. With this background and then briefly introduce current status and some highlight of degree project, then I will provide some details of the ticket of internals and last. I will talk about how to get involved and present the tclub community.

A

Okay, let's get started emerging technologies of big data, analytics machine learning, cloud and edge computing and blockchain are leading significant progress in our society, but they are also bringing confidentiality in the security issues on public cloud and blockchain.

A

Sensitive data like health and the financial records, may be exploited at runtime by untrusted computing processes running on compromised platform during in-house data exchange, configuration information across different clearance boundaries and possibly fall into the wrong hands, also not to mention the privacy issue arising offshore data supply chain and other than data privacy. uh The models algorithm they themselves also need to be well protected.

A

Once leaked attackers can steal intellectual properties or launch white box attack and easily exploit the weakness of the models fixing all this kind of risky scenarios.

A

We are in desperate need of a trusted and secure mechanism enabling us to protect both private data and computing models during marketable execution in potential and safe environment, yet preserving functionalities, performance compatibility and flexibility.

A

Flexibility, as illustrated in this figure secure computing, provides a solution for the attractions, a secure execution environment that redefines a big data business model. Even if data and model originated from different parties with no mutual trust, confidentiality and integrity can still be effectively protected.

A

Moreover, it significantly reduces uh trusted computing base and makes the whole stack easily auditable and verifiable.

A

Secure computing also called confidential. Con computing is to provide a secure place and safe place for multi-party to compute on sensitive data, trusted execution, environment or te is one of the technology for secure computing. It provides hardware-based isolation, memory, encryption and attestation, for example, intel sgx arm transom and amd scv.

A

Te implementations by different vendors in nutshell: developer need to separate the program into two parts, one in untrusted world and another in secure world or trusted world to process sensitive data within the te with all the security guarantees by te.

A

So right now, service providers like microsoft, azure, google cloud and ibm cloud have already provided the vm products on their uh cloud.

A

The goal of dkiv is to create a framework or platform that allows programmer to concentrate on business logic and automate small production of their code and data. Without worrying about technology, detail, technical detail of t development, programmers or users only need to focus on sensitive data, business logic and their interface between users and the platforms.

A

The platform manages data and executes business logic in pe computing units deploy iso distributed system.

A

So when implementing the tclav, we have several requirements for the programming languages in mind. The first one is memory safety.

A

um As we all know, a memory safety issue uh in in current application development can cause uh uh dangerous damage, but the memory safety issue in the trusted execution environment can break all security guarantees provided by hardware. Historically, we have seen that memory. Safety, vulnerabilities of te written in c and c purpose can lead to sensitive data leakage.

A

Another property we want to achieve is they are efficiency, because tea is a resource, constraint, environment and we have limited memory and io capabilities.

A

We need to have a minimal runtime and also for security. We want to have a deterministic runtime to ensure the computationality of privacy, data and integrity of code in these security properties also be remotely attested by end users.

A

So that's why we choose rust. Rust has uh has a strong type system to guarantee the memory safety of the program, so it can be statistically compiled and has a small runtime.

A

So the rust ecosystem is ready for cloud computing and have many uh third-party libraries for uh rpcs.

A

And such cloud and computing need capabilities. The community of us is also very healthy and strong to support our.

A

Development, so then, let me summarize c, clave or t apache take wave as open source or universal, secure computing platform uh retain rust, makes a computation on privacy, sensitive data safe and simple, so this project originally developed it baidu, as known as my city was open sourced in july 2019, and then we donated the uh project, along with rust, sjx sdk to apache software foundation in august 2019 and changed the project name to tclef uh 2021.

A

This year uh we also donated russ the opt trustum sdk to tclav as the subject so right now, the tk have ability to write d e in intel, sgx and amp trust zone and currently declave is under the apache incubator and the open source in the apache way. I will introduce some highlights next. You can visit our home page and the repositories to learn more.

A

Pcrave have four um basic um highlights. The first one is functionality uh to give a convenient interface for end users. Ticle is provided as a functionality service platform with many building functions. It supports tasks like machine learning, private settings, section, crypto, communication actual actual addition developer can also deploy and execute python scripts in dk. But, more importantly, unlike traditional functional service platform, decay supports both general secure computing tasks and flexible, single and multi-party, secure, computation.

A

So for security we adopt multiple uh security technologies. To enable secure computing, in particular, degree use entire cx, I'm trust zone to serve most secure, sensitive tasks with hardware based isolation, memory, encryption and the test station. Also tclab is written in rust to prevent any memory. Safety issues for usability. Degradability is components in containers.

A

uh Therefore, it can be deployed both locally and within cloud infrastructures and tickle also provide convenient and endpoint apis client sdk in a lot of languages, and also command line tools, and last modularity components in particular are designed in modular, and some components can uh like remote attestation can be easily embedded in other projects.

A

In addition, tk hdx sdk and the personal sdk can be used separate separately to write, standalone, sjs enclave and the trisome application for other purpose.

A

So since tclav is a functionality, service platform user only need to consider about functions, business logic and participants.

A

When can the user have determined about three factors? They can follow this step to execute tasks on excessive data in sgx, just like a normal function, service platform.

A

First, they register data and function to the platform and then create and run tasks. At last you can get execution results from the platform. Apis are pretty easy, we provide say rust, python and even events with client sdks.

A

Now currently, services uh are implemented in sgx enclaves and written in rust, and we have several services of in front-end core services and workers. They are authentication services, service, front-end, service storage, service management, service, scheduler, service access control and execution services.

A

We separate it into three domains and to manage the data and make sure uh sensitive data can only be stuck around inside one domain.

A

Services are communicated with rpcs and here are some interfaces defined in protobuf. I will skip this details here, but you, if you are interested in these communication interface, you can see the protobuf definitions.

A

So here's some uh here's a brief introduction of the interfaces between each service. Basically, clients communicate authenticate their id confidential credential to get a session key. This session key will be used later to communicate with the front-end service incline the then client, then register data and function if needed and assigned approve invoke task.

A

Clients can also get information of functions tasks.

A

Foreign service will redirect our valid requests to the management service and the management service get also an authorization of data, function, usage and task invocation and then persist a function. Data tags into database in the storage service and the scheduler service will fetch any uh functions.

A

Data tasks needed to be executed in the queue and dispatch tasks into execution services, so the execution service and scheduler service use subscribe and the pool model execution service can get a task and execute it.

A

After executing the function, the result will be update and the processing storage service, so a synchronizedly um clients can get a result later.

A

So that's the simple interfaces between different service. As you can see, the interface interfaces between service are pretty simple and it designed in in cloud.

A

And you can deploy this uh service in dockers and uh in the orc in your cloud infrastructure.

A

To get started with tcleaf, we provide extensive documentations, including to first try the functions: how to write functions in python and how to add, building functions, written in rust and also we provide documentations to describe of designs and implementations in details like us, our threat model and how how the mutual attestation, between in our platform and access to control modules and our build systems for rust and some other internals of implementation of tk service.

A

So if you want to read the code for each code based directories, we also have a readme to help you to get you through the code.

A

The let at last, let me give an overview of the ticker community, since tikilev is a huge project which have multiple layers. Many users can get involved in the community platform. Users can use the tk platform directly, for example, deploy the system in private infrastructure, and some other users may prefer to only use one standalone service, for example our storage service execution service and so on also a lot of project using our attestation implementations and, of course, some users directly use rust, xtx, sdk and trustor sdk to build their own applications.

A

So we are pretty open to different users to meet their various needs. Particular community have also supported many other projects like commercial products, academic research projects and some other open source projects.

A

You can see from our uh home page uh there's page called um powered by in the community. You can see uh like organizations like baidu uh by dance enigma and the projects like uh advanced, uh unified, crypto, chain uh outcome, uh and so on. So first I will not uh try to uh list them now, so they are all using tclav either the platform and the library of rust, sgx ultrasound sdk.

A

Overall, we encourage all people in the rust committee and in the sdxt community to come in and get involved.

A

So I last I want to put more information on t cliff uh if you want to follow our latest news. Please join us on the maori list and right now we have a month. We have monthly virtual meetup on zoom in every month, so follow the our monolith modeling list to see the schedule, and we also uh invite some uh speakers to talk about some topics in rust and teeclave.

A

So you can also visit our homepage to see documents and projects powered by tclav and some tutorials. You can also follow us on twitter and check out our code and at last we always call for contributions and the contributors.

A

So uh thanks so last things uh we just announced the tk trisom sdk.

A

Besides sgx sdk, we can also write just now. Applications please check out our home page. Okay, thank you. So much.