Jarek Kusmierek
Research Areas
Authored Publications
Google Publications
Other Publications
Sort By
Autopilot: Workload Autoscaling at Google Scale
Paweł Findeisen
Jacek Świderski
Przemyslaw Broniek
Beata Strack
Piotr Witusowski
Proceedings of the Fifteenth European Conference on Computer Systems, Association for Computing Machinery(2020) (to appear)
Preview abstract
In many public and private Cloud systems, users need to specify a limit for the amount of resources (CPU cores and RAM) to provision for their workloads. A job that exceeds its limits might be throttled or killed, resulting in delaying or dropping end-user requests, so human operators naturally err on the side of caution and request a larger limit than the job needs. At scale, this results in massive aggregate resource wastage.
To address this, Google uses Autopilot to configure resources automatically, adjusting both the number of concurrent tasks in a job (horizontal scaling) and the CPU/memory limits for individual tasks (vertical scaling). Autopilot walks the same fine line as human operators: its primary goal is to reduce slack – the difference between the limit and the actual resource usage – while minimizing the risk that a task is killed with an out-of-memory (OOM) error or its performance degraded because of CPU throttling. Autopilot uses machine learning algorithms applied to historical data about prior executions of a job, plus a set of finely-tuned heuristics, to walk this line. In practice, Autopiloted jobs have a slack of just 23%, compared with 46% for manually-managed jobs. Additionally, Autopilot reduces the number of jobs severely impacted by OOMs by a factor of 10.
Despite its advantages, ensuring that Autopilot was widely adopted took significant effort, including making potential recommendations easily visible to customers who had yet to opt in, automatically migrating certain categories of jobs, and adding support for custom recommenders. At the time of writing, Autopiloted jobs account for over 48% of Google's fleet-wide resource usage.
View details
Magda: A New Language for Modularity
Viviana Bono
Mauro Mulatero
Lecture Notes in Computer Science, Springer(2012), pp. 560-588
Preview abstract
We introduce Magda, a modularity-oriented programming language.
The language features lightweight mixins as units of code reuse, modular initial-
ization protocols, and a hygienic approach to identi ers. In particular, Magda's
modularity guarantees that client code of a library written in Magda will never
break as a consequence of any addition of members to the library's mixins.
View details
Modularizing Constructors
Viviana Bono
Journal of Object Technology, vol. 6, no. 9, Special Issue: TOOLS EUROPE 2007, ETH Zurich, pp. 297-317
Preview abstract
Object-oriented class-based languages provide mechanisms for the initialization of newly created objects. These mechanisms specify how an object is initialized and what information is needed to do so. The initialization protocol is usually implemented as a list of constructors. It is often the case that the initialization protocol concerns some orthogonal properties of objects. Unfortunately, if those properties have more than one option of initialization, the total number of constructors becomes exponential in the number of properties.
In this paper, we propose an alternative protocol. In our approach, instead of defining a list of constructors, it is possible to split blocks of definitions into smaller and composable pieces, in order to obtain reduction in the size of the code, better reusability, more expressiveness, and easier maintenance.
View details
