On the Value of Job Migration in Online Makespan Minimization


Makespan minimization on identical parallel machines is a classical scheduling problem. We consider the online scenario where a sequence of n jobs has to be scheduled non-preemptively on m machines so as to minimize the maximum completion time of any job. The best competitive ratio that can be achieved by deterministic online algorithms is in the range [1.88, 1.9201]. Currently no randomized online algorithm with a smaller competitiveness is known, for general m. In this paper we explore the power of job migration, i.e. an online scheduler is allowed to perform a limited number of job reassignments. Migration is a common technique used in theory and practice to balance load in parallel processing environments. As our main result we settle the performance that can be achieved by deterministic online algorithms. We develop an algorithm that is \(\alpha _m\)-competitive, for any \(m\ge 2\), where \(\alpha _m\) is the solution of a certain equation. For \(m=2\), \(\alpha _2 = 4/3\) and \(\lim _{m\rightarrow \infty } \alpha _m = W_{-1}(-1/e^2)/(1+ W_{-1}(-1/e^2)) \approx 1.4659\). Here \(W_{-1}\) is the lower branch of the Lambert W function. For \(m\ge 11\), the algorithm uses at most 7m migration operations. For smaller m, 8m to 10m operations may be performed. We complement this result by a matching lower bound: No online algorithm that uses o(n) job migrations can achieve a competitive ratio smaller than \(\alpha _m\). We finally trade performance for migrations. We give a family of algorithms that is c-competitive, for any \(5/3\le c \le 2\). For \(c= 5/3\), the strategy uses at most 4m job migrations. For \(c=1.75\), at most 2.5m migrations are used.

This is a preview of subscription content, log in to check access.

Fig. 1
Fig. 2


  1. 1.

    Aggarwal, G., Motwani, R., Zhu, A.: The load rebalancing problem. J. Algorithms 60(1), 42–59 (2006)

    MathSciNet  Article  MATH  Google Scholar 

  2. 2.

    Albers, S.: Better bounds for online scheduling. SIAM J. Comput. 29, 459–473 (1999)

    MathSciNet  Article  MATH  Google Scholar 

  3. 3.

    Bartal, Y., Karloff, H., Rabani, Y.: A better lower bound for on-line scheduling. Inf. Process. Lett. 50, 113–116 (1994)

    MathSciNet  Article  MATH  Google Scholar 

  4. 4.

    Bartal, Y., Fiat, A., Karloff, H., Vohra, R.: New algorithms for an ancient scheduling problem. J. Comput. Syst. Sci. 51, 359–366 (1995)

    MathSciNet  Article  MATH  Google Scholar 

  5. 5.

    Cesáro, E.: Sur la série harmonique. Nouvelles Ann. de Math. 3e Sér. 4, 295–296 (1885)

    MATH  Google Scholar 

  6. 6.

    Chen, B., van Vliet, A., Woeginger, G.J.: A lower bound for randomized on-line scheduling algorithms. Inf. Process. Lett. 51, 219–222 (1994)

    MathSciNet  Article  MATH  Google Scholar 

  7. 7.

    Chen, B., van Vliet, A., Woeginger, G.J.: A optimal algorithm for preemptive online scheduling. Oper. Res. Lett. 18, 127–131 (1995)

    MathSciNet  Article  MATH  Google Scholar 

  8. 8.

    Englert, M., Özmen, D., Westermann, M.: The power of reordering for online minimum makespan scheduling. SIAM J. Comput. 43(3), 1220–1237 (2014)

    MathSciNet  Article  MATH  Google Scholar 

  9. 9.

    Faigle, U., Kern, W., Turan, G.: On the performance of on-line algorithms for partition problems. Acta Cybern. 9, 107–119 (1989)

    MathSciNet  MATH  Google Scholar 

  10. 10.

    Fleischer, R., Wahl, M.: Online scheduling revisited. J. Sched. 3, 343–353 (2000)

    MathSciNet  Article  MATH  Google Scholar 

  11. 11.

    Galambos, G., Woeginger, G.: An on-line scheduling heuristic with better worst case ratio than Graham’s list scheduling. SIAM J. Comput. 22, 349–355 (1993)

    MathSciNet  Article  MATH  Google Scholar 

  12. 12.

    Graham, R.L.: Bounds for certain multi-processing anomalies. Bell Syst. Tech. J. 45, 1563–1581 (1966)

    Article  MATH  Google Scholar 

  13. 13.

    Graham, R.L.: Bounds on multiprocessing timing anomalies. SIAM J. Appl. Math. 17(2), 416–429 (1969)

    MathSciNet  Article  MATH  Google Scholar 

  14. 14.

    Gormley, T., Reingold, N., Torng, E., Westbrook, J.: Generating adversaries for request-answer games. In: Proceedings of 11th ACM-SIAM Symposium on Discrete Algorithms, pp. 564–565 (2000)

  15. 15.

    Hochbaum, D.S., Shmoys, D.B.: Using dual approximation algorithms for scheduling problems: theoretical and practical results. J. ACM 34, 144–162 (1987)

    MathSciNet  Article  Google Scholar 

  16. 16.

    Karger, D.R., Phillips, S.J., Torng, E.: A better algorithm for an ancient scheduling problem. J. Algorithms 20, 400–430 (1996)

    MathSciNet  Article  MATH  Google Scholar 

  17. 17.

    Min, X., Liu, J., Wang, Y.: Optimal semi-online algorithms for scheduling problems with reassignment on two identical machines. Inf. Process. Lett. 111(9), 423–428 (2011)

    MathSciNet  Article  MATH  Google Scholar 

  18. 18.

    Rudin, III, J.F.: Improved bounds for the on-line scheduling problem. Ph.D. Thesis. The University of Texas at Dallas, May 2001

  19. 19.

    Rudin III, J.F., Chandrasekaran, R.: Improved bounds for the online scheduling problem. SIAM J. Comput. 32, 717–735 (2003)

    MathSciNet  Article  MATH  Google Scholar 

  20. 20.

    Sanders, P., Sivadasan, N., Skutella, M.: Online scheduling with bounded migration. Math. Oper. Res. 34(2), 481–498 (2009)

    MathSciNet  Article  MATH  Google Scholar 

  21. 21.

    Sgall, J.: A lower bound for randomized on-line multiprocessor scheduling. Inf. Process. Lett. 63, 51–55 (1997)

    MathSciNet  Article  MATH  Google Scholar 

  22. 22.

    Tan, Z., Yu, S.: Online scheduling with reassignment. Oper. Res. Lett. 36(2), 250–254 (2008)

    MathSciNet  Article  MATH  Google Scholar 

  23. 23.

    Sleator, D.D., Tarjan, R.E.: Amortized efficiency of list update and paging rules. Commun. ACM 28, 202–208 (1985)

    MathSciNet  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Susanne Albers.

Additional information

A preliminary version of this paper has appeared in Proc. 20th Annual European Symposium on Algorithms (ESA), 2012. Susanne Albers’ work supported by the German Research Foundation, project Al 464/7-1.



Proof of Lemma 1

Fix \(m\ge 2\). We first evaluate \(f_m(2)\) and \(f_m(1+1/(3m))\). For \(\alpha =2\), we have \(\lceil (1-1/\alpha )m\rceil \ge m/2\). Hence \(\lceil (1-1/\alpha )m\rceil \alpha /m \ge 1\) and \(f_m(2) \ge 1\). For \(\alpha =1+ 1/(3m)\), there holds \(\lceil (1-1/\alpha )m\rceil =1\). Thus \(f_m(1+1/(3m)) = 1/(3m) H_{m-1} +1/m + 1/(3m^2)< 1/3 + 1/2 + 1/12 < 1\). It remains to show that \(f_m(\alpha )\) is continuous and strictly increasing. To this end we show that, for any \(\alpha >1\) and small \(\epsilon >0\), \(f_m(\alpha ) - f_m(\alpha -\epsilon )\) and \(f_m(\alpha +\epsilon ) - f_m(\alpha )\) converge to 0 as \(\epsilon \rightarrow 0\). Moreover \(f_m(\alpha +\epsilon ) - f_m(\alpha )\) is strictly positive.

First consider an \(\alpha >1\) such that \((1-1/\alpha )m\notin \mathbb {N}\). In this case we choose \(\epsilon > 0\) such that \(\lceil (1-1/(\alpha -\epsilon ))m\rceil = \lceil (1-1/(\alpha +\epsilon ))m\rceil = \lceil (1-1/\alpha )m\rceil \). We have

$$\begin{aligned} f_m(\alpha )= & {} (\alpha -1)(H_{m-1}-H_{\lceil (1-1/\alpha )m\rceil -1}) + \lceil (1-1/\alpha )m\rceil \alpha /m\\ f_m(\alpha -\epsilon )= & {} (\alpha -\epsilon -1)(H_{m-1}-H_{\lceil (1-1/\alpha )m\rceil -1}) + \lceil (1-1/\alpha )m\rceil (\alpha -\epsilon )/m\\ f_m(\alpha +\epsilon )= & {} (\alpha +\epsilon -1)(H_{m-1}-H_{\lceil (1-1/\alpha )m\rceil -1}) + \lceil (1-1/\alpha )m\rceil (\alpha +\epsilon )/m. \end{aligned}$$

Thus \(f_m(\alpha ) - f_m(\alpha -\epsilon ) = f_m(\alpha +\epsilon ) - f_m(\alpha ) = \epsilon (H_{m-1}-H_{\lceil (1-1/\alpha )m\rceil -1}) + \lceil (1-1/\alpha )m\rceil \epsilon /m\). Hence \(f_m(\alpha )-f_m(\alpha -\epsilon )\) and \(f_m(\alpha +\epsilon ) - f_m(\alpha )\) tend to 0 as \(\epsilon \rightarrow 0\). Since \(\alpha >1\) there holds \(\lceil (1-1/\alpha )m\rceil \ge 1\) and thus \(f_m(\alpha +\epsilon ) - f_m(\alpha ) >0\).

Next let \(\alpha >1\) such that \((1-1/\alpha )m\in \mathbb {N}\). In this case we choose \(\epsilon > 0\) such that \(\lceil (1-1/(\alpha -\epsilon ))m\rceil = \lceil (1-1/\alpha )m\rceil \) and \(\lceil (1-1/(\alpha +\epsilon ))m\rceil = \lceil (1-1/\alpha )m\rceil +1\). There holds

$$\begin{aligned} f_m(\alpha )= & {} (\alpha -1)(H_{m-1}-H_{\lceil (1-1/\alpha )m\rceil -1}) + \lceil (1-1/\alpha )m\rceil \alpha /m\\ f_m(\alpha -\epsilon )= & {} (\alpha -\epsilon -1)(H_{m-1}-H_{\lceil (1-1/\alpha )m\rceil -1}) + \lceil (1-1/\alpha )m\rceil (\alpha -\epsilon )/m\\ f_m(\alpha +\epsilon )= & {} (\alpha +\epsilon -1)(H_{m-1}-H_{\lceil (1-1/\alpha )m\rceil }) + (\lceil (1-1/\alpha )m\rceil +1) (\alpha +\epsilon )/m. \end{aligned}$$

As above \(f_m(\alpha ) - f_m(\alpha -\epsilon ) = \epsilon (H_{m-1}-H_{\lceil (1-1/\alpha )m\rceil -1}) + \lceil (1-1/\alpha )m\rceil \epsilon /m\) and the latter expression tends to 0 as \(\epsilon \rightarrow 0\). Taking into account that \((1-1/\alpha )m\in \mathbb {N}\) we obtain

$$\begin{aligned} f_m(\alpha +\epsilon ) - f_m(\alpha )= & {} -(\alpha -1)\cdot 1/((1-1/\alpha )m) + \epsilon (H_{m-1}-H_{\lceil (1-1/\alpha )m\rceil })\\&+ (\lceil (1-1/\alpha )m\rceil +1) \epsilon /m + \alpha /m\\= & {} \epsilon (H_{m-1}-H_{\lceil (1-1/\alpha )m\rceil }) + (\lceil (1-1/\alpha )m\rceil +1) \epsilon /m. \end{aligned}$$

Again, \(f_m(\alpha +\epsilon ) - f_m(\alpha )\) is strictly positive and tends to 0 as \(\epsilon \rightarrow 0\). \(\square \)

Proof of Lemma 2

We first prove that \((\alpha _m)_{m\ge 2}\) is non-decreasing. A first observation is that \(\alpha _m \le m\) because \(f_m(m)\ge 1\). We will show that, for any \(m\ge 3\) and \(1<\alpha \le m\), there holds \(f_{m-1}(\alpha ) \ge f_m(\alpha )\). This implies \(1= f_{m-1}(\alpha _{m-1}) \ge f_m(\alpha _{m-1})\). By Lemma 1, \(f_m\) is strictly increasing and thus \(\alpha _m\ge \alpha _{m-1}\). Consider a fixed \(\alpha \) with \(1<\alpha \le m\). We study two cases depending on whether or not \(\lceil (1-1/\alpha )(m-1)\rceil = \lceil (1-1/\alpha )m\rceil \).

If \(\lceil (1-1/\alpha )(m-1)\rceil = \lceil (1-1/\alpha )m\rceil \), then

$$\begin{aligned} f_m(\alpha )= & {} (\alpha -1)(H_{m-1}-H_{\lceil (1-1/\alpha )m\rceil -1}) + \lceil (1-1/\alpha )m\rceil \alpha /m\\ f_{m-1}(\alpha )= & {} (\alpha -1)(H_{m-2}-H_{\lceil (1-1/\alpha )m\rceil -1}) + \lceil (1-1/\alpha )m\rceil \alpha /(m-1). \end{aligned}$$

We obtain \(f_{m-1}(\alpha ) - f_m(\alpha ) = -(\alpha -1)/(m-1) + \lceil (1-1/\alpha )m\rceil \alpha /(m(m-1)) \ge -(\alpha -1)/(m-1) + (\alpha -1)/(m-1) =0\) and thus \(f_{m-1}(\alpha ) \ge f_m(\alpha )\).

If \(\lceil (1-1/\alpha )(m-1)\rceil < \lceil (1-1/\alpha )m\rceil \), then \(\lceil (1-1/\alpha )(m-1)\rceil = \lceil (1-1/\alpha )m\rceil -1\) and

$$\begin{aligned} f_m(\alpha )= & {} (\alpha -1)(H_{m-1}-H_{\lceil (1-1/\alpha )m\rceil -1}) + \lceil (1-1/\alpha )m\rceil \alpha /m\\ f_{m-1}(\alpha )= & {} (\alpha -1)(H_{m-2}-H_{\lceil (1-1/\alpha )m\rceil -2}) + (\lceil (1-1/\alpha )m\rceil -1) \alpha /(m-1). \end{aligned}$$

Since \(\alpha >1\) there holds \(\lceil (1-1/\alpha )(m-1)\rceil \ge 1\). Hence in our case \(\lceil (1-1/\alpha )m\rceil \ge 2\) and \(\lceil (1-1/\alpha )m\rceil -1 >0\). We obtain

$$\begin{aligned} \textstyle {f_{m-1}(\alpha ) - f_m(\alpha ) = -{\alpha -1\over m-1} + {\alpha -1\over \lceil (1-1/\alpha )m\rceil -1} + \lceil (1-1/\alpha )m\rceil {\alpha \over m(m-1)} - {\alpha \over m-1}.} \end{aligned}$$

Choose x, with \(0\le x <1\), such that \(\lceil (1-1/\alpha )m\rceil = (1-1/\alpha )m +x\). Then

$$\begin{aligned} \textstyle {f_{m-1}(\alpha ) - f_m(\alpha )}= & {} \textstyle { -{\alpha -1\over m-1} + {\alpha -1\over (1-1/\alpha )m +x-1} + (1-1/\alpha )m {\alpha \over m(m-1)} + {\alpha x\over m(m-1)}- {\alpha \over m-1}}\\= & {} \textstyle {{\alpha -1\over (1-1/\alpha )m +x-1} + {\alpha x\over m(m-1)}- {\alpha \over m-1}} \end{aligned}$$

In order to establish \(f_{m-1}(\alpha ) - f_m(\alpha )\ge 0\) is suffices to show

$$\begin{aligned} \textstyle {{\alpha -1\over (1-1/\alpha )m +x-1} \ge {\alpha (m-x)\over m(m-1)}.} \end{aligned}$$

This is equivalent to \((\alpha -1)m(m-1) \ge (m-x)((\alpha -1)m +\alpha x-\alpha )\). Standard algebraic manipulation yield that this is equivalent to \(m \ge mx - \alpha x^2+\alpha x\). Let \(g(x) = mx - \alpha x^2+\alpha x\), for any real number x. This function is increasing for any \(x < (m+\alpha )/(2\alpha )\). Since \(\alpha \le m\), the function is increasing for any \(x <1\). As \(g(0) = 0\) and \(g(1) = m\), it follows that \(m \ge mx - \alpha x^2+\alpha x\) holds for all \(0\le x <1\). We conclude \(f_{m-1}(\alpha ) - f_m(\alpha )\ge 0\).

It is easy to verify that \(f_2(4/3)=1\). We show that \(\lim _{m\rightarrow \infty } \alpha _m\) is upper bounded by \(W_{-1}(-1/e^2)/(1+ W_{-1}(-1/e^2))\). Cesáro [5] proved

$$\begin{aligned} 0< H_m - \frac{1}{2} \ln \left( m(m+1) \right) -\gamma < \frac{1}{6m(m+1)}, \end{aligned}$$

where \(\gamma \approx 0.577\) is the Euler-Mascheroni constant. Using this inequality we find, for any c with \(0< c\le 1\) and \(\lceil cm\rceil -2 >0\),

$$\begin{aligned} H_{m-1} - H_{\lceil cm \rceil -2 }> & {} \frac{1}{2} \ln ((m-1)m) + \gamma - \frac{1}{2} \ln ((\lceil cm \rceil -2)(\lceil cm\rceil -1))\\&-\gamma - \frac{1}{6(\lceil cm \rceil -2)(\lceil cm \rceil -1)}\\\ge & {} \frac{1}{2} \left( \ln (m-1)+ \ln m - \ln (cm -1) - \ln ( cm) \right) - \frac{1}{2(\lceil cm \rceil -1)} \\= & {} \frac{1}{2} \left( \ln (m-1)+ \ln m - \ln (c(m-1/c)) - \ln (cm) \right) - \frac{1}{2(\lceil cm \rceil -1)} \\= & {} \frac{1}{2} \left( \ln (m-1) - \ln (m-1/c) -2 \ln (c) \right) - \frac{1}{2(\lceil cm \rceil -1)} \\\ge & {} \frac{1}{2} \left( 2 \ln (1/c) \right) - \frac{1}{2(\lceil cm \rceil -1)} \\\ge & {} \ln (1/c) - \frac{1}{2(cm-1)}, \end{aligned}$$

where the second to last inequality holds since \(\ln (m-1/c) \le \ln (m-1)\). for \(0<c\le 1\) and sufficiently large m. We obtain

$$\begin{aligned} f_m(\alpha )= & {} (\alpha -1)(H_{m-1}-H_{\lceil (1-1/\alpha )m \rceil -1}) + \left( \lceil (1-1/\alpha )m \rceil \right) \frac{\alpha }{m} \\> & {} (\alpha -1)\left( \ln (\frac{\alpha }{\alpha -1}) - \frac{1}{2((1-1/\alpha )m -1)} - \frac{1}{\lceil (1-1/\alpha )m \rceil -1} \right) \\&+ \left( \lceil (1-1/\alpha )m \rceil \right) \frac{\alpha }{m} \\\ge & {} (\alpha -1)\left( \ln \left( \frac{\alpha }{\alpha -1}\right) - \frac{1}{(1-1/\alpha )m -1} \right) + \alpha -1 =: F(m). \end{aligned}$$

Obviously, \(\lim _{m\rightarrow \infty } F(m) = (\alpha -1) \ln (\frac{\alpha }{\alpha -1}) +\alpha -1\). We show that \((\alpha -1) \ln (\frac{\alpha }{\alpha -1}) +\alpha -1 = 1\), for \(\alpha =\frac{1}{1-\delta }\), where \(\delta = -1/W_{-1}(-1/e^2)\).

Equation \((\alpha -1) \ln (\frac{\alpha }{\alpha -1}) + \alpha -1 = 1\) is equivalent to \(\ln (\frac{\alpha }{\alpha -1})+1 = \frac{1}{\alpha -1}\), which in turn is equivalent to

$$\begin{aligned} \frac{\alpha }{\alpha -1} \cdot e = e^\frac{1}{\alpha -1}. \end{aligned}$$

Substituting \(x=1/(\alpha -1)\), which is equivalent to \(\alpha =1/x+1\), we find that the above is equivalent to \(xe+e = e^x\). Applying the Lambert W function we find that \(x=-W_{-1}(-1/e^2)-1\) is a solution of the former equality. Substituting we conclude that in fact \(\alpha = W_{-1}(-1/e^2)/(1+ W_{-1}(-1/e^2))\) satisfies the equality. Using the same techniques we can show that \(\lim _{m\rightarrow \infty } \alpha _m\) is lower bounded by \(W_{-1}(-1/e^2)/(1+ W_{-1}(-1/e^2))\). In the calculations, (1) yields that \(H_{m-1} - H_{\lceil cm \rceil } < \ln (1/c) + 1/(2m)\). \(\square \)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Albers, S., Hellwig, M. On the Value of Job Migration in Online Makespan Minimization. Algorithmica 79, 598–623 (2017). https://doi.org/10.1007/s00453-016-0209-9

Download citation


  • Scheduling
  • Makespan minimization
  • Online algorithm
  • Competitive analysis
  • Job migration