Do Le Quoc

Diana Petrescu, Arsany Guirguis, Do Le Quoc et al.

Storage disaggregation underlies today's cloud and is naturally complemented by pushing down some computation to storage, thus mitigating the potential network bottleneck between the storage and compute tiers. We show how ML training benefits from storage pushdowns by focusing on transfer learning (TL), the widespread technique that democratizes ML by reusing existing knowledge on related tasks. We propose HAPI, a new TL processing system centered around two complementary techniques that address challenges introduced by disaggregation. First, applications must carefully balance execution across tiers for performance. HAPI judiciously splits the TL computation during the feature extraction phase yielding pushdowns that not only improve network time but also improve total TL training time by overlapping the execution of consecutive training iterations across tiers. Second, operators want resource efficiency from the storage-side computational resources. HAPI employs storage-side batch size adaptation allowing increased storage-side pushdown concurrency without affecting training accuracy. HAPI yields up to 2.5x training speed-up while choosing in 86.8% of cases the best performing split point or one that is at most 5% off from the best.

Robert Krahn, Donald Dragoti, Franz Gregor et al.

Trusted Execution Environments (TEEs), such as Intel Software Guard eXtensions (SGX), are considered as a promising approach to resolve security challenges in clouds. TEEs protect the confidentiality and integrity of application code and data even against privileged attackers with root and physical access by providing an isolated secure memory area, i.e., enclaves. The security guarantees are provided by the CPU, thus even if system software is compromised, the attacker can never access the enclave's content. While this approach ensures strong security guarantees for applications, it also introduces a considerable runtime overhead in part by the limited availability of protected memory (enclave page cache). Currently, only a limited number of performance measurement tools for TEE-based applications exist and none offer performance monitoring and analysis during runtime. This paper presents TEEMon, the first continuous performance monitoring and analysis tool for TEE-based applications. TEEMon provides not only fine-grained performance metrics during runtime, but also assists the analysis of identifying causes of performance bottlenecks, e.g., excessive system calls. Our approach smoothly integrates with existing open-source tools (e.g., Prometheus or Grafana) towards a holistic monitoring solution, particularly optimized for systems deployed through Docker containers or Kubernetes and offers several dedicated metrics and visualizations. Our evaluation shows that TEEMon's overhead ranges from 5% to 17%.

Do Le Quoc, Christof Fetzer

Federated Learning (FL) is an emerging machine learning paradigm that enables multiple clients to jointly train a model to take benefits from diverse datasets from the clients without sharing their local training datasets. FL helps reduce data privacy risks. Unfortunately, FL still exist several issues regarding privacy and security. First, it is possible to leak sensitive information from the shared training parameters. Second, malicious clients can collude with each other to steal data, models from regular clients or corrupt the global training model. To tackle these challenges, we propose SecFL - a confidential federated learning framework that leverages Trusted Execution Environments (TEEs). SecFL performs the global and local training inside TEE enclaves to ensure the confidentiality and integrity of the computations against powerful adversaries with privileged access. SecFL provides a transparent remote attestation mechanism, relying on the remote attestation provided by TEEs, to allow clients to attest the global training computation as well as the local training computation of each other. Thus, all malicious clients can be detected using the remote attestation mechanisms.

Wojciech Ozga, Do Le Quoc, Christof Fetzer

Trust is of paramount concern for tenants to deploy their security-sensitive services in the cloud. The integrity of VMs in which these services are deployed needs to be ensured even in the presence of powerful adversaries with administrative access to the cloud. Traditional approaches for solving this challenge leverage trusted computing techniques, e.g., vTPM, or hardware CPU extensions, e.g., AMD SEV. But, they are vulnerable to powerful adversaries, or they provide only load time (not runtime) integrity measurements of VMs. We propose WELES, a protocol allowing tenants to establish and maintain trust in VM runtime integrity of software and its configuration. WELES is transparent to the VM configuration and setup. It performs an implicit attestation of VMs during a secure login and binds the VM integrity state with the secure connection. Our prototype's evaluation shows that WELES is practical and incurs low performance overhead.

Wojciech Ozga, Do Le Quoc, Christof Fetzer

Confidential multi-stakeholder machine learning (ML) allows multiple parties to perform collaborative data analytics while not revealing their intellectual property, such as ML source code, model, or datasets. State-of-the-art solutions based on homomorphic encryption incur a large performance overhead. Hardware-based solutions, such as trusted execution environments (TEEs), significantly improve the performance in inference computations but still suffer from low performance in training computations, e.g., deep neural networks model training, because of limited availability of protected memory and lack of GPU support. To address this problem, we designed and implemented Perun, a framework for confidential multi-stakeholder machine learning that allows users to make a trade-off between security and performance. Perun executes ML training on hardware accelerators (e.g., GPU) while providing security guarantees using trusted computing technologies, such as trusted platform module and integrity measurement architecture. Less compute-intensive workloads, such as inference, execute only inside TEE, thus at a lower trusted computing base. The evaluation shows that during the ML training on CIFAR-10 and real-world medical datasets, Perun achieved a 161x to 1560x speedup compared to a pure TEE-based approach.

Do Le Quoc, Franz Gregor, Sergei Arnautov et al.

Data-driven intelligent applications in modern online services have become ubiquitous. These applications are usually hosted in the untrusted cloud computing infrastructure. This poses significant security risks since these applications rely on applying machine learning algorithms on large datasets which may contain private and sensitive information. To tackle this challenge, we designed secureTF, a distributed secure machine learning framework based on Tensorflow for the untrusted cloud infrastructure. secureTF is a generic platform to support unmodified TensorFlow applications, while providing end-to-end security for the input data, ML model, and application code. secureTF is built from ground-up based on the security properties provided by Trusted Execution Environments (TEEs). However, it extends the trust of a volatile memory region (or secure enclave) provided by the single node TEE to secure a distributed infrastructure required for supporting unmodified stateful machine learning applications running in the cloud. The paper reports on our experiences about the system design choices and the system deployment in production use-cases. We conclude with the lessons learned based on the limitations of our commercially available platform, and discuss open research problems for the future work.

Wojciech Ozga, Do Le Quoc, Christof Fetzer

Trusted computing defines how to securely measure, store, and verify the integrity of software controlling a computer. One of the major challenges that make them hard to be applied in practice is the issue with software updates. Specifically, an operating system update causes the integrity violation because it changes the well-known initial state trusted by remote verifiers, such as integrity monitoring systems. Consequently, the integrity monitoring of remote computers becomes unreliable due to the high amount of false positives. We address this problem by adding an extra level of indirection between the operating system and software repositories. We propose a trusted software repository (TSR), a secure proxy that overcomes the shortcomings of previous approaches by sanitizing software packages. Sanitization consists of modifying unsafe installation scripts and adding digital signatures in a way software packages can be installed in the operating system without violating its integrity. TSR leverages shielded execution, i.e., Intel SGX, to achieve confidentiality and integrity guarantees of the sanitization process. TSR is transparent to package managers, and requires no changes in the software packages building and distributing processes. Our evaluation shows that running TSR inside SGX is practical; since it induces only ~1.18X performance overhead during package sanitization compared to the native execution without SGX. TSR supports 99.76% of packages available in the main and community repositories of Alpine Linux while increasing the total repository size by 3.6%.

Franz Gregor, Wojciech Ozga, Sébastien Vaucher et al.

Trust is arguably the most important challenge for critical services both deployed as well as accessed remotely over the network. These systems are exposed to a wide diversity of threats, ranging from bugs to exploits, active attacks, rogue operators, or simply careless administrators. To protect such applications, one needs to guarantee that they are properly configured and securely provisioned with the "secrets" (e.g., encryption keys) necessary to preserve not only the confidentiality, integrity and freshness of their data but also their code. Furthermore, these secrets should not be kept under the control of a single stakeholder - which might be compromised and would represent a single point of failure - and they must be protected across software versions in the sense that attackers cannot get access to them via malicious updates. Traditional approaches for solving these challenges often use ad hoc techniques and ultimately rely on a hardware security module (HSM) as root of trust. We propose a more powerful and generic approach to trust management that instead relies on trusted execution environments (TEEs) and a set of stakeholders as root of trust. Our system, PALAEMON, can operate as a managed service deployed in an untrusted environment, i.e., one can delegate its operations to an untrusted cloud provider with the guarantee that data will remain confidential despite not trusting any individual human (even with root access) nor system software. PALAEMON addresses in a secure, efficient and cost-effective way five main challenges faced when developing trusted networked applications and services. Our evaluation on a range of benchmarks and real applications shows that PALAEMON performs efficiently and can protect secrets of services without any change to their source code.

Roland Kunkel, Do Le Quoc, Franz Gregor et al.

Machine learning has become a critical component of modern data-driven online services. Typically, the training phase of machine learning techniques requires to process large-scale datasets which may contain private and sensitive information of customers. This imposes significant security risks since modern online services rely on cloud computing to store and process the sensitive data. In the untrusted computing infrastructure, security is becoming a paramount concern since the customers need to trust the thirdparty cloud provider. Unfortunately, this trust has been violated multiple times in the past. To overcome the potential security risks in the cloud, we answer the following research question: how to enable secure executions of machine learning computations in the untrusted infrastructure? To achieve this goal, we propose a hardware-assisted approach based on Trusted Execution Environments (TEEs), specifically Intel SGX, to enable secure execution of the machine learning computations over the private and sensitive datasets. More specifically, we propose a generic and secure machine learning framework based on Tensorflow, which enables secure execution of existing applications on the commodity untrusted infrastructure. In particular, we have built our system called TensorSCONE from ground-up by integrating TensorFlow with SCONE, a shielded execution framework based on Intel SGX. The main challenge of this work is to overcome the architectural limitations of Intel SGX in the context of building a secure TensorFlow system. Our evaluation shows that we achieve reasonable performance overheads while providing strong security properties with low TCB.

Do Le Quoc, Martin Beck, Pramod Bhatotia et al.

How to preserve users' privacy while supporting high-utility analytics for low-latency stream processing? To answer this question: we describe the design, implementation, and evaluation of PRIVAPPROX, a data analytics system for privacy-preserving stream processing. PRIVAPPROX provides three properties: (i) Privacy: zero-knowledge privacy guarantees for users, a privacy bound tighter than the state-of-the-art differential privacy; (ii) Utility: an interface for data analysts to systematically explore the trade-offs between the output accuracy (with error-estimation) and query execution budget; (iii) Latency: near real-time stream processing based on a scalable "synchronization-free" distributed architecture. The key idea behind our approach is to marry two existing techniques together: namely, sampling (used in the context of approximate computing) and randomized response (used in the context of privacy-preserving analytics). The resulting marriage is complementary - it achieves stronger privacy guarantees and also improves performance, a necessary ingredient for achieving low-latency stream analytics.