Archive for June 2006

Don't panic, it's just a performance review

June 26, 2006

I’m about to send out the appraisal form, which we are also using for self-appraisals this year. I need them back by July 7th, so there’s not a lot of time.

Feel free to give yourself ratings, as well as make comments that I’ll find helpful in writing your appraisal. Feel free to write something separate, too.

The format and layout of the appraisal tool, and the numerical rating system, were dictated by HR and the senior management of the Institute. As a result, you may find that the final rating seems lower than you’d expect. Don’t panic.

First, differences of something like ten or fifteen points aren’t meaningful. There are five broad categories, and I don’t expect there will be anybody in the top category of the bottom two. Within the remaining two categories, while a “Meets and Exceeds” is better than a “Successfully Meets”, differences of several “points” within “Successfully Meets” probably has more to do with the math behind the instrument than real ratings differences.

Second, the evaluation is against expectations. If you’re a Principal or a Senior, that means expectations are already very high, so the ratings are done against very high expectations. For everybody, the expectation is that you’re going to meet the civility standards, so it would take unusual deviation from expectations (either positive or negative) to get anything but a “3” rating on your first goal. The professional development goals are specific to each of you, so we set specific expectations. I, for example, expect to get a “2” on my PD goal, because I only did part of it, and never got around to getting my PMI certification.

We’ll be sending the exact same form to your lead (and if you’re a lead, your CE) and in some cases to Mission Office people. If there are people you’d like to have me ask for input, please let me know.

But mostly, don’t panic.



Cool/amusing nanopeople

June 19, 2006

I just thought this was cool, and rather amusing.

Kind of like smileys on the nanoscale.


Performance Appraisal Process

June 19, 2006

Yep, HR has finalized the plan and schedule for Performance Appraisals. The process kicks off this week, and gets us doing the appraisals right after the end of the goal period. (Which is certainly what *I* want to see.)

The process this year is intended to work as follows: (lifted shamelessly from their memo)

Week of June 19, 2006

  • Division Management should meet with all Division reviewing managers and establish
    the performance appraisal process and deadlines for this year.
  • Managers should inform employees of the beginning of the process and provide
    instructions for completion of self-appraisal, if required.
  • Managers should solicit input from other raters.
  • Manager training for performance appraisal process conducted.
  • Managers should begin compiling information to be used for input to performance
    appraisals. Managers may begin completing performance appraisal forms for employes.

Weeks of June 26 and July 3

  • Employees may begin accessing the appraisal form to complete self-appraisal comments.
  • Employees should be instructed to indicate self-appraisal comments. Employees or other
    raters are not to provide rating.
  • Managers complete performance appraisal forms for employees.
  • Managers should be collecting other rater input. Other raters must sign off on input.

Week of July 10

  • Self-appraisals, if used, due to managers no later than COB, July 7.
  • Managers complete performance appraisals

Week of July 17 and Week of July 24

  • Managers complete performance appraisals

Week of July 31

  • All appraisals due to Division Management.
  • Division Management reviews appraisals and forwards to HR review no later than COB
    August 4.

Weeks of August 7 and 14

  • HR completes appraisal review and returns signed appraisals to Division for employee
    performance conferences.
  • Once reviewed appraisal is returned, managers may meet with employees one on one for
    individual performance conferences.

Weeks of August 21 and August 28

  • Merit Increase Process – Worksheets delivered to Divisions. Must be returned to HR
    with recommendations by September 1.

Weeks of September 4 and September 11

  • Executive Review of Merit Increases

Week of September 18

  • Communication of Merit Increases

October 1, 2006

  • New pay rates in effect first pay period beginning in October.

Performance appraisals; raises

June 14, 2006

We’re still waiting for HR to finalize their plans for performance appraisals this year. They are redoing the form and the rating system, and will be redoing the goals themselves next year, so I expect both pushback and changes.

Raises are, of course, determined in large part by the appraisals. The escalation the past few years has been right around 3.5%, which means the raise pool can’t be much different from 3.5%. I don’t expect this year to be any different. Pete Stockman yesterday mentioned that the JWST contract was flat, without any escalation, so we’ll have to reduce the work to make even “cost of living” increases possible.

The CPI this year is shaping up to quite a bit higher than the last few years. This morning BLS put out the monthly CPI updates. The CPI-U is up 4.3%, and there are disturbing trends in the underlying data. The transportation index, for example, is up 9%, and the energy index is up a whopping 23.6%. For the Northeast (where I live) the overall index is higher — 4.9%.

I don’t see an immediate change in the CPI, and people I know who actually do economics are worried about a return to “stagflation” of the sort seen in the late 70s and early 80s. That would be bad. Hopefully Bernanke has some good ideas about avoiding that problem.


Last item on the All-Hands. Retreat and Ivan.

June 13, 2006

Joe’s comments on the retreat:

End of FY05 and start of FY06 was very challenging for the management team, which got in the way of working together. We met offsite in March to start a recovery process. We put a bunch of issues on the table, and have been talking about them since.

We now have a better understanding of our expectations and re-learned how to work together as a team.

Who is Ivan?:

Management consultant with a technical background who has worked with Matt before at Gemini, as well as JPL, Keck and ALMA.

Joe expected Ivan to have a little low-key discussion about how to “tune-up” ESS. Turned in to a workshop with users and customers and ESS staff to talk about what our problems are, and how we might structure itself to best serve the needs of STScI.

You can get a good description of what we did at the Theme Teams wiki pages.


Notes from the ESS All-Hands.

June 13, 2006

I didn’t realize it had been most of a week since I last posted. Yikes.

Here’s the usual brain dump.

Flash news – Tropical Storm Alberto has come ashore.

Joe started with acknowledgement: Tony Krueger’s AURA award, Christine Ritchie and Tony Rodgers for the PC and preprocessor software; Rob Hawkins, Brian Sietz and Christine for the SPIKE GUI. Warren Hack and Perry Greenfield for work on JWST test work; Alan Welty for taking over HST commanding; JWST OSS Team for their work on JWST commanding; and Lisa Gardner for the PROMPT revival work. Lee Quick has her second BS in Info Systems from JHU. And finally, Mike Bielefeld is leaving at the end of June.

Some announcements: Shelley Marshall has been here a month and a half. Warren Miller is in HSTMO working on the HLA project; we’ll likely post a new CE position. The AURA Voluntary Early Retirement program is available 6/15 to 8/31. You have to be 55 or 57 and be hired before 1/1/2001.

Financials: We’re out of the woods for FY06; still no news on the HST contract, just rumours. We’re not expecting any budget change for FY07.

The Picnic is Friday.


Pete Stockman gave a talk on JWST status.

NGST has completed a high-bay for observatory integration – the OTE will be integrated at Goddard and then mated to the spacecraft in the new bay. The 18 beryllium mirror blanks are all complete and being polished. Two are down to final polishing stage. They’ve also done a demonstration of a detector to get the mirror segments aligned to within a micron. That gets us to the starting point for our wavefront alignment system.

The supports for the mirror segments are even more complicated than Peter understood, but that’s necessary to keep from distorting the mirror segments itself. That integration is stsrting. The other thing that’s getting ready is the cryotank for thermal-vac testing. The decision to do the testing with the mirror facing up (rather than down) is a big cost and time saver — 3months and $100 million dollars.

The hardware for the Backplane Stability test is all fabricated – Warren and Perry are working on the software – and that will go into a chamber at Marshall to show that we can do the metrology on the mirror backplane. The frame gets hung on wires, and the goal is to make sure that the frame expands and contracts only in-plane. If the frame distorts out-of-plane, it means we can’t focus the main mirror, which would be A Bad Thing. This is one of the Technology Readiness Level-6 milestones, which is key for getting through the Non-Advocate Review in January.

Next year we’ll do the rest of the mirror fine machining, do the sunshield folding model test, and a bunch of CDRs for things like the fast steering mirror.

It’s a big mirror composed of 18 segments with a nice hold in the middle. Each segment is independently steered and focused on a rigid (we hope) backplane.

All the instruments have passed PDR. NIRCam has a bunch of hardware already done, NIRSpec is coming along and have picked their flight detector candidates. NIRSpec’s shutter array is a set of tiny little doors (365 by 171 elements) to pick objects to take spectra of. Very cool, and part of the ITAR restriction rationale. GGS and MIRI are starting to build test hardware.

At STScI: We’ve had two PRDS tools, FITSWriter, Point Spread Function software, work on onboard commanding scripts, calibration plans, system architecture support, etc. Spent a bunch of time talking about the 4D work to support the Marshall test. Trying to take out 100 angstrom vibration and take huge amounts of data (order of 14TBytes) to look at the backplane components.

FY07 funding is the same, but most people got a small raise, so we’re doing more with less. There are a bunch of things to do in FY07, including planning for the start of real SOC development in FY08. The big system engineering things are interfaces between subsystems, and to the outside world, and some small-scale operations concepts. FY08 is supposed to be a 30 person staff-up.

Carl went through his talk on the Network and Storage Upgrade project. I’m not going to reiterate what I wrote before, just some comments on the changes.

The tie on Carl’s door means Joe isn’t allowed to bother him. This project is what Carl has been doing for the last six months. The goal is to get network and storage upgraded to 1Gbps to the desktop and central storage. Carl’s goal: Show Me The Money.

The old model worked okay (even well) when we were pushing small amounts (many megabytes or so) around. When we’re doubling every 18 months the model falls apart.

Why does data volume increase faster than archive volume? Because we’re doing more with the data – mosaicing, drizzle, associations – that we keep intermediate products for and make multiple products with. The network will be (is?) the showstopper for doing collaborative work.

People who worked on Carl’s tests agreed that the new centralized model would work if there was a scaleable solution, both storage and network performance. The solution has to be used by everybody (pretty much) to be effective.

Muller building will go 1 Gb to the desktop; Bloomberg will not upgrade.

Backup is a big driver for both central storage and information lifecycle management.

For fixed desktops, the sharing model is pretty well udnerstood. For mobile users like me, it’s still not understood real well.


Carl on Storage and Network Enhancements

June 8, 2006

Carl is using the SAS staff to preview his all-hands talk on storage. He’ll be giving the same talk (with edits based on our feedback) at next Tuesday’s all-hands.

To get the extra money to redo the network and storage infrastructure, we had to demonstrate that we needed to fund a major infrastructure initiative.
Current infrastructure will not support future storage growth at the Institute.

Current model is direct attached storage with limited backups. 40 TB on 1300 cross-mounted disks. 10% of network load is simple keep-alive cross mounting messaging. No accountability for data integrity, significant hardware and network dependency.

The data volume growth is driven by both science and functional work. That will grow from routine work, and will be exploded by SM-4. 60% growth per month; doubling every 18 months. 260 TB by 2010.

Carl described how Paul described how INS does an ACS image of M51. All the steps in the pipeline, mosaicing, drizzling, combination. All that consumes lots of space, and grows the space. As instruments/detector packages grow, so will our storage needs, and the network and disk space will break down.

So he walked through some scenarios about the desired end-state. Main goal is to significantly improve data integrity without losing performance.
First example was centralized storage over a gigabit network. The example was clearly adequate, if it will scale. This still only works, however, if people move from direct attached to centralized storage.

In order to do storage, we first need to improve network bandwidth. The goal is 10Gbit backbone, 1Gbit to desktop, but no upgrade to Bloomberg. That will stay at 10/100 to the desktop, and about 1Gb backbone. In Muller, three closet swtiches talk to a central switch over 10Gb.

For storage, we have a central store with virtualization and ILM, so that the desktop talks over a lan to virtual disks that are part of a central store. Data migrates from the NAS to a disk library, to a tape backup. All the migration off the NAS is over the SAN fabric; only talking to clients happens over the NAS. Virtualization takes care of where things are, ILM gives you the ability to define rules about where things should live. The rules are configurable, so we can be flexible based on users, file types, age, etc.

Current costs are based on engineering estimates, no attempt to negotiate. Cable costs have gone way up, so network is between $411K and $611K. (One third to one half of which is copper in petroleum-based insulators.) The total cost is $2.4M of which $1.78M is for initial procurement in FY06, though cost was split across FY06 and FY07.

Partly because of the organizational changes, Carl will continue to lead the process for a while.

And Carl will do a verison of this presentation at the All-Hands next week.