Low Overhead Hardware-Assisted Virtual Machine Analysis and Profiling
Bibliographic record
Abstract
Cloud infrastructure providers need reliable performance analysis tools for their nodes. Moreover, the analysis of Virtual Machines (VMs) is a major requirement in quantifying cloud performance. However, root cause analysis, in case of unexpected crashes or anomalous behavior in VMs, remains a major challenge. Modern tracing tools such as LTTng allow fine grained analysis - albeit at a minimal execution overhead, and being OS dependent. In this paper, we propose HAVAna, a hardware-assisted VM analysis algorithm that gathers and analyzes pure hardware trace data, without any dependence on the underlying OS or performance analysis infrastructure. Our approach is totally non-intrusive and does not require any performance statistics, trace or log gathering from the VM. We used the recently introduced Intel PT ISA extensions on modern Intel Skylake processors to demonstrate its efficiency and observed that, in our experimental scenarios, it leads to a tiny overhead of up to 1%, as compared to 3.6-28.7% for similar VM trace analysis done with software-only schemes such as LTTng. Our proposed VM trace analysis algorithm has also been open-sourced for further enhancements and to the benefit of other developers. Furthermore, we developed interactive Resource and Process Control Flow visualization tools to analyze the hardware trace data and present a real-life usecase in the paper that allowed us to see unexpected resource consumption by VMs.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".