AXaaS: Case for acceleration as a service

Powers, Nathaniel; Ailing, Alexander; Gyampoh‐Vidogah, Regina; Soyata, Tolga

doi:10.1109/glocomw.2014.7063416

Cited by 13 publications

(3 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The top of this figure (i.e., the data transformation path) shows the transformations that the incoming data stream [ ] must go through to produce a properlyformatted ciphertext. This top part is assumed to be performed during data acquisition by a computationally capable device, such as the patient's smartphone or a cloudlet (Wang, Liu, & Soyata, 2014;Powers, Alling, Gyampoh-Vidogah, & Soyata, 2014) (see Figure 1). The bottom of Figure 8 (i.e., the functional transformation path) will be the focus of this section, which details the steps that must be taken to convert the computation ( (. ))…”

Section: Fhe Implementation Stepsmentioning

confidence: 99%

Towards Privacy-Preserving Medical Cloud Computing Using Homomorphic Encryption

Kocabaş

Soyata

Advances in Wireless Technologies and Telecommunication

Self Cite

View full text Add to dashboard Cite

Personal health monitoring tools, such as commercially available wireless ECG patches, can significantly reduce healthcare costs by allowing patient monitoring outside the healthcare organizations. These tools transmit the acquired medical data into the cloud, which could provide an invaluable diagnosis tool for healthcare professionals. Despite the potential of such systems to revolutionize the medical field, the adoption of medical cloud computing in general has been slow due to the strict privacy regulations on patient health information. We present a novel medical cloud computing approach that eliminates privacy concerns associated with the cloud provider. Our approach capitalizes on Fully Homomorphic Encryption (FHE), which enables computations on private health information without actually observing the underlying data. For a feasibility study, we present a working implementation of a long-term cardiac health monitoring application using a well-established open source FHE library.

show abstract

Section: Fhe Implementation Stepsmentioning

confidence: 99%

Towards Privacy-Preserving Medical Cloud Computing Using Homomorphic Encryption

Kocabaş

Soyata

Advances in Wireless Technologies and Telecommunication

Self Cite

View full text Add to dashboard Cite

show abstract

“…Therefore, the data has to be transferred to a nearby computationally capable device. We conceptualize this device to be either a smart-phone, with approximately 5 GFLOPS computational capability (iPhone5s, n.d.) or a cloudlet with close to 100 GFLOPS capability (Soyata, Ba, Heinzelman, Kwon, & Shi, 2013;Wang, Liu, & Soyata, 2014;Alling, Powers, & Soyata, 2015;Powers, Alling, Gyampoh-Vidogah, & Soyata, 2014). Encrypted signals should include two different flavors: a traditional encryption, such as PGP or AES (NIST-AES, 2001), or FHE.…”

Section: Motivating Applicationmentioning

confidence: 99%

Secure Health Monitoring in the Cloud Using Homomorphic Encryption

Ames

Venkitasubramaniam

Page

et al.

Advances in Wireless Technologies and Telecommunication

Self Cite

View full text Add to dashboard Cite

Extending cloud computing to medical software, where the hospitals rent the software from the provider sounds like a natural evolution for cloud computing. One problem with cloud computing, though, is ensuring the medical data privacy in applications such as long term health monitoring. Previously proposed solutions based on Fully Homomorphic Encryption (FHE) completely eliminate privacy concerns, but are extremely slow to be practical. Our key proposition in this paper is a new approach to applying FHE into the data that is stored in the cloud. Instead of using the existing circuit-based programming models, we propose a solution based on Branching Programs. While this restricts the type of data elements that FHE can be applied to, it achieves dramatic speed-up as compared to traditional circuit-based methods. Our claims are proven with simulations applied to real ECG data.

show abstract

“…"As a service" (aaS) offerings provide convenient, configurable solutions to computational problems by allocating resources over the internet (Dsouza, Kabbedjik, Seo, Jansen, & Brinkkemper, 2012;Costa, Migliavacca, Pietzuch, & Wolf, 2012;Sandikkaya & Harmanci, 2012). To run an exciting new breed of resource-intensive mobile applications such as Real-Time Face Recognition (Powers, Alling, Gyampoh-Vidogah, & Soyata, 2014), Augmented Reality, Real-time Language Translation (Google-Translate; MS-Translator), and Surveillance, it is possible for a user to rent cloud instances. However, the steps required for such a process will discourage even the most willing user.…”

Section: Background and Motivationmentioning

confidence: 99%

Selling FLOPs

Powers

Soyata

Advances in Wireless Technologies and Telecommunication

Self Cite

View full text Add to dashboard Cite

To meet the user demand for an ever-increasing mobile-cloud computing performance for resource-intensive mobile applications, we propose a new service architecture called Acceleration as a Service (AXaaS). We formulate AXaaS based on the observation that most resource-intensive applications, such as real-time face-recognition and augmented reality, have similar resource-demand characteristics: a vast majority of the program execution time is spent on a limited set of library calls, such as Generalized Matrix-Multiply operations (GEMM), or FFT. Our AXaaS model suggests accelerating only these operations by the Telecom Service Providers (TSP). We envision the TSP offering this service through a monthly computational service charge, much like their existing monthly bandwidth charge. We demonstrate the technological and business feasibility of AXaaS on a proof-of-concept real-time face recognition application. We elaborate on the consumer, developer, and the TSP view of this model. Our results confirm AXaaS as a novel and viable business model.

show abstract

AXaaS: Case for acceleration as a service

Cited by 13 publications

References 14 publications

Towards Privacy-Preserving Medical Cloud Computing Using Homomorphic Encryption

Towards Privacy-Preserving Medical Cloud Computing Using Homomorphic Encryption

Secure Health Monitoring in the Cloud Using Homomorphic Encryption

Selling FLOPs

Contact Info

Product

Resources

About