S-FaaS: Trustworthy and Accountable Function-as-a-Service using Intel SGX

被引:35
作者
Alder, Fritz [1 ,3 ]
Asokan, N. [2 ,3 ]
Kurnikov, Arseny [3 ,4 ]
Paverd, Andrew [3 ,5 ]
Steiner, Michael [6 ]
机构
[1] Katholieke Univ Leuven, Imec DistriNet, Leuven, Belgium
[2] Univ Waterloo, Waterloo, ON, Canada
[3] Aalto Univ, Espoo, Finland
[4] Ericsson, Stockholm, Sweden
[5] Microsoft Res, Redmond, WA USA
[6] Intel Labs, Hillsboro, OR USA
来源
CCSW'19: PROCEEDINGS OF THE 2019 ACM SIGSAC CONFERENCE ON CLOUD COMPUTING SECURITY WORKSHOP | 2019年
关键词
Function-as-a-Service; Resource Measurement; Intel SGX;
D O I
10.1145/3338466.3358916
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
Function-as-a-Service (FaaS) is a recent and popular cloud computing paradigm in which the function provider specifies a function to be run and is billed only for the computational resources used by that function. Compared to other cloud paradigms, FaaS requires significantly more fine-grained measurement of functions' compute time and memory usage. Since functions are short and stateless, small ephemeral entities (e.g. individuals or underutilized data centers) can become FaaS service providers. However, this exacerbates the already substantial challenges of 1) ensuring integrity of computation, 2) minimizing information revealed to the service provider, and 3) accurately measuring computational resource usage. To address these challenges, we introduce S-FaaS, the first architecture and implementation of FaaS to provide strong security and accountability guarantees using Intel SGX. To match the dynamic event-driven nature of FaaS, we introduce a new key distribution enclave and a novel transitive attestation protocol. A core contribution of S-FaaS is our set of reusable resource measurement mechanisms that securely measure compute time and memory usage inside an enclave. We have integrated S-FaaS into the OpenWhisk FaaS framework and provide this as open source software.
引用
收藏
页码:185 / 199
页数:15
相关论文
共 44 条
[1]  
Amazon Web Services, 2018, AWS EC2 SPOT PRIC
[2]  
Amazon Web Services, 2018, AWS LAMBD PRIC
[3]  
[Anonymous], 2019, ANKR NETWORK
[4]  
[Anonymous], 2018, CLIMATEPREDICTION NE
[5]  
[Anonymous], 2018, GOLEM NETWORK
[6]  
[Anonymous], 2016, ConvNetJS
[7]  
[Anonymous], 2017, P 1 INT WORKSH SEC D
[8]  
[Anonymous], 2019, INTEL TSX
[9]  
[Anonymous], 2017, 26 USENIX SEC S
[10]  
[Anonymous], 2016, P 1 WORKSH SYST SOFT