Scott Shenker
YOU?
Author Swipe
View article: Making Cellular Networks More Efficient By Roaming-in-Place
Making Cellular Networks More Efficient By Roaming-in-Place Open
We propose Roaming-in-Place (RinP), a technique for dynamically sharing capacity across mobile network operators. RinP is a new form of infrastructure sharing that expands the traditional notion of roaming in cellular networks such that us…
View article: Bandwidth Allocation for Cloud-Augmented Autonomous Driving
Bandwidth Allocation for Cloud-Augmented Autonomous Driving Open
Autonomous vehicle (AV) control systems increasingly rely on ML models for tasks such as perception and planning. Current practice is to run these models on the car's local hardware due to real-time latency constraints and reliability conc…
View article: Extracting Database Access-control Policies From Web Applications
Extracting Database Access-control Policies From Web Applications Open
To safeguard sensitive user data, web developers typically rely on implicit access-control policies, which they implement using access checks and query filters. This ad hoc approach is error-prone as these scattered checks and filters are …
View article: A Decidable Case of Query Determinacy: Project-Select Views
A Decidable Case of Query Determinacy: Project-Select Views Open
Query determinacy is decidable for project-select views and a project-select-join query with no self joins, as long as the selection predicates are in a first-order theory for which satisfiability is decidable.
View article: Revisiting Cache Freshness for Emerging Real-Time Applications
Revisiting Cache Freshness for Emerging Real-Time Applications Open
Caching is widely used in industry to improve application performance by\nreducing data-access latency and taking the load off the backend\ninfrastructure. TTLs have become the de-facto mechanism used to keep cached\ndata reasonably fresh …
View article: SkyServe: Serving AI Models across Regions and Clouds with Spot Instances
SkyServe: Serving AI Models across Regions and Clouds with Spot Instances Open
Recent years have witnessed an explosive growth of AI models. The high cost of hosting AI services on GPUs and their demanding service requirements, make it timely and challenging to lower service costs and guarantee service quality. While…
View article: Managing Bandwidth: The Key to Cloud-Assisted Autonomous Driving
Managing Bandwidth: The Key to Cloud-Assisted Autonomous Driving Open
Prevailing wisdom asserts that one cannot rely on the cloud for critical real-time control systems like self-driving cars. We argue that we can, and must. Following the trends of increasing model sizes, improvements in hardware, and evolvi…
View article: Principles for Internet Congestion Management
Principles for Internet Congestion Management Open
Given the technical flaws with - -and the increasing non-observance of - -the TCP-friendliness paradigm, we must rethink how the Internet should manage bandwidth allocation. We explore this question from first principles, but remain within…
View article: Efficient Microsecond-scale Blind Scheduling with Tiny Quanta
Efficient Microsecond-scale Blind Scheduling with Tiny Quanta Open
A longstanding performance challenge in datacenter-based applications is how to efficiently handle incoming client requests that spawn many very short (μs scale) jobs that must be handled with high throughput and low tail latency. When no …
View article: How I Learned to Stop Worrying About CCA Contention
How I Learned to Stop Worrying About CCA Contention Open
This paper asks whether inter-flow contention between congestion control algorithms (CCAs) is a dominant factor in determining a flow's bandwidth allocation in today's Internet. We hypothesize that CCA contention typically does not determi…
View article: Bringing Reconfigurability to the Network Stack
Bringing Reconfigurability to the Network Stack Open
Reconfiguring the network stack allows applications to specialize the implementations of communication libraries depending on where they run, the requests they serve, and the performance they need to provide. Specializing applications in t…
View article: Out of Hand for Hardware? Within Reach for Software!
Out of Hand for Hardware? Within Reach for Software! Open
Events that take 10s to 100s of ns like cache misses increasingly cause CPU stalls. However, hiding the latency of these events is challenging: hardware mechanisms suffer from the lack of flexibility, whereas prior software mechanisms fall…
View article: Access Control for Database Applications: Beyond Policy Enforcement
Access Control for Database Applications: Beyond Policy Enforcement Open
There have been many recent advances in enforcing finegrained access control for database-backed applications. However, operators face significant challenges both before and after an enforcement mechanism has been deployed. We identify thr…
View article: Extracting the Essential Simplicity of the Internet
Extracting the Essential Simplicity of the Internet Open
Looking past inessential complexities to explain the Internet's simple yet daring design.
View article: The case for an internet primitive for fault localization
The case for an internet primitive for fault localization Open
Modern distributed applications run across numerous microservices and components deployed in cloud datacenters, using shared cloud services for computing and storage, edge services such as content distribution networks, network functions s…
View article: Global content revocation on the internet
Global content revocation on the internet Open
Common wisdom holds that once personal content such as photographs have been shared on the Internet, they will stay there forever. This paper explores how we could allow users to reclaim some degree of their privacy by "revoking" previousl…
View article: 3PO: Programmed Far-Memory Prefetching for Oblivious Applications
3PO: Programmed Far-Memory Prefetching for Oblivious Applications Open
Using memory located on remote machines, or far memory, as a swap space is a promising approach to meet the increasing memory demands of modern datacenter applications. Operating systems have long relied on prefetchers to mask the increase…
View article: The Sky Above The Clouds
The Sky Above The Clouds Open
Technology ecosystems often undergo significant transformations as they mature. For example, telephony, the Internet, and PCs all started with a single provider, but in the United States each is now served by a competitive market that uses…
View article: Blockaid: Data Access Policy Enforcement for Web Applications
Blockaid: Data Access Policy Enforcement for Web Applications Open
Modern web applications serve large amounts of sensitive user data, access to which is typically governed by data-access policies. Enforcing such policies is crucial to preventing improper data access, and prior work has proposed many enfo…
View article: Answering three questions about networking research
Answering three questions about networking research Open
Researchers often talk about specific technical trends or research topics. But we rarely talk about how and why we do the research that we do. The process of submitting and reviewing papers puts our ideas through a particular kind of filte…
View article: Democratizing cellular access with CellBricks
Democratizing cellular access with CellBricks Open
Markets in which competition thrives are good for both consumers and innovation but, unfortunately, competition is not thriving in the increasingly important cellular market. We propose CellBricks, a novel cellular architecture that lowers…
View article: AutoTune: Improving End-to-end Performance and Resource Efficiency for Microservice Applications
AutoTune: Improving End-to-end Performance and Resource Efficiency for Microservice Applications Open
Most large web-scale applications are now built by composing collections (from a few up to 100s or 1000s) of microservices. Operators need to decide how many resources are allocated to each microservice, and these allocations can have a la…
View article: From cloud computing to sky computing
From cloud computing to sky computing Open
We consider the future of cloud computing and ask how we might guide it towards a more coherent service we call sky computing. The barriers are more economic than technical, and we propose reciprocal peering as a key enabling step.
View article: Revitalizing the public internet by making it extensible
Revitalizing the public internet by making it extensible Open
There is now a significant and growing functional gap between the public Internet, whose basic architecture has remained unchanged for several decades, and a new generation of more sophisticated private networks. To address this increasing…