Users of decentralized finance (DeFi) applications face significant risks from adversarial actions that manipulate the order of transactions to extract value from users. Such actions -- an adversarial form of what is called maximal-extractable value (MEV) -- impact both individual outcomes and the stability of the DeFi ecosystem. MEV exploitation, moreover, is being institutionalized through an architectural paradigm known Proposer-Builder Separation (PBS). This work introduces a system called PROF (PRotected Order Flow) that is designed to limit harmful forms of MEV in existing PBS systems. PROF aims at this goal using two ideas. First, PROF imposes an ordering on a set ("bundle") of privately input transactions and enforces that ordering all the way through to block production -- preventing transaction-order manipulation. Second, PROF creates bundles whose inclusion is profitable to block producers, thereby ensuring that bundles see timely inclusion in blocks. PROF is backward-compatible, meaning that it works with existing and future PBS designs. PROF is also compatible with any desired algorithm for ordering transactions within a PROF bundle (e.g., first-come, first-serve, fee-based, etc.). It executes efficiently, i.e., with low latency, and requires no additional trust assumptions among PBS entities. We quantitatively and qualitatively analyze incentive structure of PROF, and its utility to users compared with existing solutions. We also report on inclusion likelihood of PROF transactions, and concrete latency numbers through our end-to-end implementation.

An optimistic rollup (ORU) scales a blockchain’s throughput by delegating computation to an untrusted remote chain (L2), refereeing any state claim disagreements between mutually distrusting L2 operators via an interactive dispute resolution protocol. State-of-the-art ORUs employ a monolithic dispute resolution protocol that tightly couples an L1 referee with a specific L2 client binary—oblivious to the system’s higher-level semantics. We argue that this approach (1) magnifies monoculture failure risk, by precluding trust-minimized and permissionless participation using operator-chosen client software; (2) leads to an unnecessarily large and difficult-to-audit TCB; and, (3) suffers from a frequently-triggered, yet opaque upgrade process—both further increasing auditing overhead, and broadening the governance attack surface.To address these concerns, we outline a methodology for designing a secure and resilient ORU with a minimal TCB, by facilitating opportunistic 1-of-N-version programming. Due to its unique challenges and opportunities, we ground this work concretely in the context of the Ethereum ecosystem—where ORUs have gained significant traction. Specifically, we design a semantically-aware proof system, natively targeting the EVM and its instruction set. We present an implementation in a new ORU, Specular, that opportunistically leverages Ethereum’s existing client diversity with minimal source modification, demonstrating our approach’s feasibility.

Hyperparameter tuning is a necessary step in training and deploying machine learning models. Most prior work on hyperparameter tuning has studied methods for maximizing model accuracy under a time constraint, assuming a fixed cluster size. While this is appropriate in data center environments, the increased deployment of machine learning workloads in cloud settings necessitates studying hyperparameter tuning with an elastic cluster size and time and monetary budgets. While recent work has leveraged the elasticity of the cloud to minimize the execution cost of a pre-determined hyperparameter tuning job originally designed for fixed-cluster sizes, they do not aim to maximize accuracy.

Hyperparameter tuning is essential to achieving state-of-the-art accuracy in machine learning (ML), but requires substantial compute resources to perform. Existing systems primarily focus on effectively allocating resources for a hyperparameter tuning job under fixed resource constraints. We show that the available parallelism in such jobs changes dynamically over the course of execution and, therefore, presents an opportunity to leverage the elasticity of the cloud.

As cluster computing frameworks such as Spark, Dryad, Flink, and Ray are being deployed in mission critical applications and on larger and larger clusters, their ability to tolerate failures is growing in importance. These frameworks employ two broad approaches for fault tolerance: checkpointing and lineage. Checkpointing exhibits low overhead during normal operation but high overhead during recovery, while lineage-based solutions make the opposite tradeoff.