article thumbnail

The Return of the Frame Pointers

Brendan Gregg

The problem is that this system has a default libc that has been compiled without frame pointers, so any stack walking stops at the libc layer, producing a partial stack that's missing the application frames. This is pretty common and usually goes unnoticed as the flame graph looks ok at first glance. Just to name a couple of languages.

Java 145
article thumbnail

Windows Timer Resolution: The Great Rule Change

Randon ASCII

The behavior of the Windows scheduler changed significantly in Windows 10 2004, in a way that will break a few applications, and there appears to have been no announcement, and the documentation has not been updated. First, a bit of operating-system design context. The answer is hardware interrupts.

Insiders

Sign Up for our Newsletter

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

Trending Sources

article thumbnail

File systems unfit as distributed storage backends: lessons from ten years of Ceph evolution

The Morning Paper

File systems unfit as distributed storage backends: lessons from 10 years of Ceph evolution Aghayev et al., In this case, the assumption that a distributed storage backend should clearly be layered on top of a local file system. A distributed file system provides a unified view over aggregated storage from multiple physical machines.

Storage 64
article thumbnail

What Adrian Did Next?—?Part 2?—?Sun Microsystems

Adrian Cockcroft

Photo by Adrian I spent six years at Cambridge Consultants, building some interesting systems, managing our Sun workstations and learning a lot, but by then Sun had opened a sales office across the street, and I wanted to find out what they were going to release next, before everyone else. as a result there were product and messaging changes.

Tuning 52
article thumbnail

DynamoDB One Year Later - All Things Distributed

All Things Distributed

Werner Vogels weblog on building scalable and robust distributed systems. This allows us to tune both our hardware and our software to ensure that the end-to-end service is both cost-efficient and highly performant. All Things Distributed. DynamoDB One Year Later: Bigger, Better, and 85% Cheaperâ?¦. Comments ().

Ecommerce 139
article thumbnail

Data Redundancy With the PostgreSQL Citus Extension

Percona

Depending on the configuration, one can tune a hardware RAID for either performance or redundancy. Can you think of a better system that can stay up without losing time initiating failovers when a node fails? Now let’s stretch our imagination and consider a second method of high availability, ala Citus.

C++ 78
article thumbnail

Welcome to the Jungle

Sutter's Mill

Now welcome to the hardware jungle. For the first time in the history of computing, mainstream hardware is no longer a single-processor von Neumann machine, and never will be again. The free lunch is over. In 2005, however, mainstream computing hit a wall. There’s no going back. That was the first act. I hope you enjoy it.