Good News 4/3/23

Prof Ming Liu and colleagues’ ASPLOS ‘23 paper, “A Generic Service to Provide In-Network Aggregation for Key-Value Streams”, was selected as a Distinguished Paper. Great job! [1]

Prof Kirthi Kandasamy and colleagues had a paper conditionally accepted into OSDI ‘23, entitled “Cilantro: A Framework for Performance-Aware Resource Allocation for General Objectives via Online Feedback”. Great work! [2]

Two graduate students in our department won NSF graduate fellowships: Hailey Johnson (advised by Prof Bilge Mutlu) and Ivan Hu (advised by Prof Dieter van Melkebeek). Congratulations to all of you! [3]

Prof Bilge Mutlu and colleagues had two further papers accepted into the ACM Interaction Design and Children (IDC) Conference, entitled “My Unconditional Homework Buddy:” Designing Augmented Homework Experiences for Children with a Social Learning Companion Robot” and “Family Theories in Child-Robot Interactions: Understanding Families as a Whole for Child-Robot Interaction Design”. Well done! [4]


[1] See for details.

Paper title: A Generic Service to Provide In-Network Aggregation for Key-Value Streams

Authors: Yongchao He, Wenfei Wu, Yanfang Le, Ming Liu, ChonLam Lao

Abstract: Key-value stream aggregation is a common operation in distributed systems, which requires intensive computation and network resources. We propose a generic in-network aggregation service for key-value streams, ASK, to accelerate the aggregation operations in diverse distributed applications. ASK is a switch-host co-designed system, where the programmable switch provides a best-effort aggregation service, and the host runs a daemon to interact with applications. ASK makes in-depth optimization tailored to traffic characteristics, hardware restrictions, and network unreliable natures: it vectorizes multiple key-value tuples’ aggregation of one packet in one switch pipeline pass, which improves the per-host’s goodput; it develops a lightweight reliability mechanism for key-value stream’s asynchronous aggregation, which guarantees computation correctness; it designs a hot-key agnostic prioritization for key-skewed workloads, which improves the switch memory utilization. We prototype ASK and use it to support Spark and BytePS. The evaluation shows that ASK could accelerate pure key-value aggregation tasks by up to 155 times and big data jobs by 3-5 times, and be backward compatible with existing INA-empowered distributed training solutions with the same speedup.

[2] Cilantro: A Framework for Performance-Aware Resource Allocation for General Objectives via Online Feedback

Authors: R. Bhardwaj*, K. Kandasamy*, A. Biswal, W. Guo, B. Hindman, J. Gonzalez, M. Jordan, I. Stoica

Abstract: Traditional systems for allocating finite cluster resources among competing jobs have either aimed at providing fairness, relied on users to specify their resource requirements, or have estimated these requirements via surrogate metrics (e.g. CPU utilization). These approaches do not account for a job’s real world performance (e.g. P95 latency). Existing performance-aware systems use offline profiled data and/or are designed for specific allocation objectives. In this work, we argue that resource allocation systems should directly account for real-world performance and the varied allocation objectives of users. In this pursuit, we build Cilantro.

At the core of Cilantro is an online learning mechanism which forms feedback loops with the jobs to estimate the resource to performance mappings and load shifts. This relieves users from the onerous task of job profiling and collects reliable real-time feedback. This is then used to achieve a variety of user-specified scheduling objectives. Cilantro handles the uncertainty in the learned models by adapting the underlying policy to work with confidence bounds. We demonstrate this in two settings. First, in a multi-tenant 1000-CPU cluster with 20 independent jobs, three of Cilantro’s policies outperform 9 other baselines on three different performance-aware scheduling objectives, improving user utilities by up to 1.2 − 3.7x. Second, in a microservices setting, where 160 CPUs must be distributed between 19 inter-dependent microservices, Cilantro outperforms 3 other baselines, reducing the end-to-end P99 latency to x0.57 the next best baseline.


The NSF Graduate Research Fellowship Program (GRFP) helps ensure the vitality of the human resource base of science and engineering in the United States and reinforces its diversity. The program recognizes and supports outstanding graduate students in NSF-supported science, technology, engineering, and mathematics disciplines who are pursuing research-based master’s and doctoral degrees at accredited United States institutions.

As the oldest graduate fellowship of its kind, the GRFP has a long history of selecting recipients who achieve high levels of success in their future academic and professional careers. The reputation of the GRFP follows recipients and often helps them become life-long leaders that contribute significantly to both scientific innovation and teaching. Past fellows include numerous Nobel Prize winners, former U.S. Secretary of Energy, Steven Chu, Google founder, Sergey Brin and Freakonomics co-author, Steven Levitt.

Fellows share in the prestige and opportunities that become available when they are selected. Fellowships provide the student with a three-year annual stipend of $37,000 along with a $12,000 cost of education allowance for tuition and fees (paid to the institution), as well as access to opportunities for professional development available to NSF-supported graduate students. Fellowships may only be used for an eligible graduate degree program at an academic institution accredited in, and having a campus located in, the US, its territories, possessions, or the Commonwealth of Puerto Rico.

NSF Fellows are anticipated to become knowledge experts who can contribute significantly to research, teaching, and innovations in science and engineering. These individuals are crucial to maintaining and advancing the nation’s technological infrastructure and national security as well as contributing to the economic well-being of society at large.

So that the nation can build fully upon the strength and creativity of a diverse society, the Foundation welcomes applications from all qualified individuals. Women, under-represented minorities and people with disabilities are encouraged to apply.

The fellowship is competitive, and those planning to apply should devote a sincere effort to their application. See the Applicants section for more information.

[4] Bengisu Cagiltay, Bilge Mutlu, Joseph Michaelis, “My Unconditional Homework Buddy: Designing Augmented Homework Experiences for Children with a Social Learning Companion Robot”

We aim to design robotic educational support systems that can promote socially and intellectually meaningful learning experiences for students while they complete school work outside of class. To pursue this goal, we conducted participatory design studies with 10 children (aged 10–12) to explore their design needs for robot-assisted homework. We investigated children’s current ways of doing homework, the type of support they receive while doing homework, and co-designed the speech and expressiveness of a homework companion robot. Children and parents attending our design sessions explained that an emotionally expressive social robot as a homework aid can support students’ motivation and engagement, as well as their affective state. Children primarily perceived the robot as a dedicated assistant at home, capable of forming meaningful friendships, or a shared classroom learning resource. We present key design recommendations for augmenting students’ homework experiences with a learning companion robot to support their educational experiences.

Bengisu Cagiltay, Bilge Mutlu, Margaret Kerr, “Family Theories in Child-Robot Interactions: Understanding Families as a Whole for Child-Robot Interaction Design”

In this work we discuss a theoretically motivated family-centered design approach for child-robot interactions, adapted by Family Systems Theory (FST) and Family Ecological Model (FEM). Long-term engagement and acceptance of robots in the home is influenced by factors that surround the child and family, such as child-sibling-parent relationships, family routines, rituals, and values. A family-centered approach to interaction design is essential when developing in-home technology for children, especially for social agents like robots that can form relationships and connections with them. We review related literature in family theories and connect it with child-robot interaction and child-computer interaction research. We present two case studies that exemplify how family theories, FST and FEM, can inform how robots might be integrated into homes when applied to research in child-robot and family-robot interaction design. Finally, we pose five overarching recommendations for a family centered design approach in child-robot interactions.