Authenticated key-value stores with hardware enclaves

Kai Li; Yuzhe Tang; Qi Zhang; Jianliang Xu; Ju Chen

doi:10.1145/3491084.3491425

Authenticated key-value stores with hardware enclaves

Kai Li, Yuzhe Tang, Qi Zhang, Jianliang Xu, Ju Chen

Research output: Chapter in Book/Entry/Poem › Conference contribution

3 Scopus citations

Abstract

Authenticated data storage on an untrusted platform is an important computing paradigm for cloud applications ranging from data outsourcing, to cryptocurrency and general transparency logs. These modern applications increasingly feature update-intensive workloads, whereas existing authenticated data structures (ADSs) designed with in-place updates are inefficient to handle such workloads. This work addresses the issue and presents a novel authenticated log-structured merge tree (eLSM) based key-value store built on Intel SGX. We present a system design that runs the code of eLSM store inside enclave. To circumvent the limited enclave memory (128 MB with the latest Intel CPUs), we propose to place the memory buffer of the eLSM store outside the enclave and protect the buffer using a new authenticated data structure by digesting individual LSM-tree levels. We design protocols to support data integrity, (range) query completeness, and freshness. Our protocol causes small proofs by including the Merkle proofs at selected levels. We implement eLSM on top of Google LevelDB and Facebook RocksDB with minimal code change and performance interference. We evaluate the performance of eLSM under the YCSB workload benchmark and show a performance advantage of up to 4.5X speedup.

Original language	English (US)
Title of host publication	Middleware 2021 Industry Track - Proceedings of the 2021 International Middleware Conference Industrial Track
Publisher	Association for Computing Machinery, Inc
Pages	1-8
Number of pages	8
ISBN (Electronic)	9781450391528
DOIs	https://doi.org/10.1145/3491084.3491425
State	Published - Dec 6 2021
Event	22nd International Middleware Conference, Middleware 2021 - Virtual, Online, Canada Duration: Dec 6 2021 → Dec 10 2021

Publication series

Name	Middleware 2021 Industry Track - Proceedings of the 2021 International Middleware Conference Industrial Track

Conference

Conference	22nd International Middleware Conference, Middleware 2021
Country/Territory	Canada
City	Virtual, Online
Period	12/6/21 → 12/10/21

Keywords

LSM trees
SGX
data integrity
enclave
key-value stores
query authentication
storage consistency

ASJC Scopus subject areas

Information Systems
Software

Access to Document

10.1145/3491084.3491425

Cite this

Li, K., Tang, Y., Zhang, Q., Xu, J., & Chen, J. (2021). Authenticated key-value stores with hardware enclaves. In Middleware 2021 Industry Track - Proceedings of the 2021 International Middleware Conference Industrial Track (pp. 1-8). (Middleware 2021 Industry Track - Proceedings of the 2021 International Middleware Conference Industrial Track). Association for Computing Machinery, Inc. https://doi.org/10.1145/3491084.3491425

Authenticated key-value stores with hardware enclaves. / Li, Kai; Tang, Yuzhe; Zhang, Qi et al.
Middleware 2021 Industry Track - Proceedings of the 2021 International Middleware Conference Industrial Track. Association for Computing Machinery, Inc, 2021. p. 1-8 (Middleware 2021 Industry Track - Proceedings of the 2021 International Middleware Conference Industrial Track).

Research output: Chapter in Book/Entry/Poem › Conference contribution

Li, K, Tang, Y, Zhang, Q, Xu, J & Chen, J 2021, Authenticated key-value stores with hardware enclaves. in Middleware 2021 Industry Track - Proceedings of the 2021 International Middleware Conference Industrial Track. Middleware 2021 Industry Track - Proceedings of the 2021 International Middleware Conference Industrial Track, Association for Computing Machinery, Inc, pp. 1-8, 22nd International Middleware Conference, Middleware 2021, Virtual, Online, Canada, 12/6/21. https://doi.org/10.1145/3491084.3491425

Li K, Tang Y, Zhang Q, Xu J, Chen J. Authenticated key-value stores with hardware enclaves. In Middleware 2021 Industry Track - Proceedings of the 2021 International Middleware Conference Industrial Track. Association for Computing Machinery, Inc. 2021. p. 1-8. (Middleware 2021 Industry Track - Proceedings of the 2021 International Middleware Conference Industrial Track). doi: 10.1145/3491084.3491425

Li, Kai ; Tang, Yuzhe ; Zhang, Qi et al. / Authenticated key-value stores with hardware enclaves. Middleware 2021 Industry Track - Proceedings of the 2021 International Middleware Conference Industrial Track. Association for Computing Machinery, Inc, 2021. pp. 1-8 (Middleware 2021 Industry Track - Proceedings of the 2021 International Middleware Conference Industrial Track).

@inproceedings{785816e8ac7e48c0be62b6cce6bbca5a,

title = "Authenticated key-value stores with hardware enclaves",

abstract = "Authenticated data storage on an untrusted platform is an important computing paradigm for cloud applications ranging from data outsourcing, to cryptocurrency and general transparency logs. These modern applications increasingly feature update-intensive workloads, whereas existing authenticated data structures (ADSs) designed with in-place updates are inefficient to handle such workloads. This work addresses the issue and presents a novel authenticated log-structured merge tree (eLSM) based key-value store built on Intel SGX. We present a system design that runs the code of eLSM store inside enclave. To circumvent the limited enclave memory (128 MB with the latest Intel CPUs), we propose to place the memory buffer of the eLSM store outside the enclave and protect the buffer using a new authenticated data structure by digesting individual LSM-tree levels. We design protocols to support data integrity, (range) query completeness, and freshness. Our protocol causes small proofs by including the Merkle proofs at selected levels. We implement eLSM on top of Google LevelDB and Facebook RocksDB with minimal code change and performance interference. We evaluate the performance of eLSM under the YCSB workload benchmark and show a performance advantage of up to 4.5X speedup.",

keywords = "LSM trees, SGX, data integrity, enclave, key-value stores, query authentication, storage consistency",

author = "Kai Li and Yuzhe Tang and Qi Zhang and Jianliang Xu and Ju Chen",

note = "Publisher Copyright: {\textcopyright} 2021 ACM.; 22nd International Middleware Conference, Middleware 2021 ; Conference date: 06-12-2021 Through 10-12-2021",

year = "2021",

month = dec,

day = "6",

doi = "10.1145/3491084.3491425",

language = "English (US)",

series = "Middleware 2021 Industry Track - Proceedings of the 2021 International Middleware Conference Industrial Track",

publisher = "Association for Computing Machinery, Inc",

pages = "1--8",

booktitle = "Middleware 2021 Industry Track - Proceedings of the 2021 International Middleware Conference Industrial Track",

}

TY - GEN

T1 - Authenticated key-value stores with hardware enclaves

AU - Li, Kai

AU - Tang, Yuzhe

AU - Zhang, Qi

AU - Xu, Jianliang

AU - Chen, Ju

PY - 2021/12/6

Y1 - 2021/12/6

N2 - Authenticated data storage on an untrusted platform is an important computing paradigm for cloud applications ranging from data outsourcing, to cryptocurrency and general transparency logs. These modern applications increasingly feature update-intensive workloads, whereas existing authenticated data structures (ADSs) designed with in-place updates are inefficient to handle such workloads. This work addresses the issue and presents a novel authenticated log-structured merge tree (eLSM) based key-value store built on Intel SGX. We present a system design that runs the code of eLSM store inside enclave. To circumvent the limited enclave memory (128 MB with the latest Intel CPUs), we propose to place the memory buffer of the eLSM store outside the enclave and protect the buffer using a new authenticated data structure by digesting individual LSM-tree levels. We design protocols to support data integrity, (range) query completeness, and freshness. Our protocol causes small proofs by including the Merkle proofs at selected levels. We implement eLSM on top of Google LevelDB and Facebook RocksDB with minimal code change and performance interference. We evaluate the performance of eLSM under the YCSB workload benchmark and show a performance advantage of up to 4.5X speedup.

AB - Authenticated data storage on an untrusted platform is an important computing paradigm for cloud applications ranging from data outsourcing, to cryptocurrency and general transparency logs. These modern applications increasingly feature update-intensive workloads, whereas existing authenticated data structures (ADSs) designed with in-place updates are inefficient to handle such workloads. This work addresses the issue and presents a novel authenticated log-structured merge tree (eLSM) based key-value store built on Intel SGX. We present a system design that runs the code of eLSM store inside enclave. To circumvent the limited enclave memory (128 MB with the latest Intel CPUs), we propose to place the memory buffer of the eLSM store outside the enclave and protect the buffer using a new authenticated data structure by digesting individual LSM-tree levels. We design protocols to support data integrity, (range) query completeness, and freshness. Our protocol causes small proofs by including the Merkle proofs at selected levels. We implement eLSM on top of Google LevelDB and Facebook RocksDB with minimal code change and performance interference. We evaluate the performance of eLSM under the YCSB workload benchmark and show a performance advantage of up to 4.5X speedup.

KW - LSM trees

KW - SGX

KW - data integrity

KW - enclave

KW - key-value stores

KW - query authentication

KW - storage consistency

UR - http://www.scopus.com/inward/record.url?scp=85121471601&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85121471601&partnerID=8YFLogxK

U2 - 10.1145/3491084.3491425

DO - 10.1145/3491084.3491425

M3 - Conference contribution

AN - SCOPUS:85121471601

T3 - Middleware 2021 Industry Track - Proceedings of the 2021 International Middleware Conference Industrial Track

SP - 1

EP - 8

BT - Middleware 2021 Industry Track - Proceedings of the 2021 International Middleware Conference Industrial Track

PB - Association for Computing Machinery, Inc

T2 - 22nd International Middleware Conference, Middleware 2021

Y2 - 6 December 2021 through 10 December 2021

ER -

Authenticated key-value stores with hardware enclaves

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this